2026-02-23

Title: BioBridge: Bridging Proteins and Language for Enhanced Biological Reasoning with LLMs

• Authors: Yujia Wang, Jihong Guan, Wengen Li, Shuigeng Zhou, Xuhong Wang
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17680
• Pdf URL: https://arxiv.org/pdf/2602.17680
• Copy Paste: [[2602.17680]] BioBridge: Bridging Proteins and Language for Enhanced Biological Reasoning with LLMs(https://arxiv.org/abs/2602.17680)
• Keywords: large language model
• Abstract: Existing Protein Language Models (PLMs) often suffer from limited adaptability to multiple tasks and exhibit poor generalization across diverse biological contexts. In contrast, general-purpose Large Language Models (LLMs) lack the capability to interpret protein sequences and fall short in domain-specific knowledge, limiting their capacity for effective biosemantic reasoning. To combine the advantages of both, we propose BioBridge, a domain-adaptive continual pretraining framework for protein understanding. This framework employs Domain-Incremental Continual Pre-training (DICP) to infuse protein domain knowledge and general reasoning corpus into a LLM simultaneously, effectively mitigating catastrophic forgetting. Cross-modal alignment is achieved via a PLM-Projector-LLM pipeline, which maps protein sequence embeddings into the semantic space of the language model. Ultimately, an end-to-end optimization is adopted to uniformly support various tasks, including protein property prediction and knowledge question-answering. Our proposed BioBridge demonstrates performance comparable to that of mainstream PLMs on multiple protein benchmarks, such as EC and BindingDB. It also achieves results on par with LLMs on general understanding tasks like MMLU and RACE. This showcases its innovative advantage of combining domain-specific adaptability with general-purpose language competency.

Title: LATMiX: Learnable Affine Transformations for Microscaling Quantization of LLMs

• Authors: Ofir Gordon, Lior Dikstein, Arnon Netzer, Idan Achituve, Hai Victor Habi
• Subjects: cs.LG, cs.CL
• Abstract URL: https://arxiv.org/abs/2602.17681
• Pdf URL: https://arxiv.org/pdf/2602.17681
• Copy Paste: [[2602.17681]] LATMiX: Learnable Affine Transformations for Microscaling Quantization of LLMs(https://arxiv.org/abs/2602.17681)
• Keywords: robust, large language model
• Abstract: Post-training quantization (PTQ) is a widely used approach for reducing the memory and compute costs of large language models (LLMs). Recent studies have shown that applying invertible transformations to activations can significantly improve quantization robustness by reducing activation outliers; however, existing approaches are largely restricted to rotation or Hadamard-based transformations. Moreover, most studies focused primarily on traditional quantization schemes, whereas modern hardware increasingly supports the microscaling (MX) data format. Attempts to combine both showed severe performance degradation, leading prior work to introduce assumptions on the transformations. In this work, we take a complementary perspective. First, we provide a theoretical analysis of transformations under MX quantization by deriving a bound on the quantization error. Our analysis emphasizes the importance of accounting for both the activation distribution and the underlying quantization structure. Building on this analysis, we propose LATMiX, a method that generalizes outlier reduction to learnable invertible affine transformations optimized using standard deep learning tools. Experiments show consistent improvements in average accuracy for MX low-bit quantization over strong baselines on a wide range of zero-shot benchmarks, across multiple model sizes.

Title: Duality Models: An Embarrassingly Simple One-step Generation Paradigm

• Authors: Peng Sun, Xinyi Shang, Tao Lin, Zhiqiang Shen
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17682
• Pdf URL: https://arxiv.org/pdf/2602.17682
• Copy Paste: [[2602.17682]] Duality Models: An Embarrassingly Simple One-step Generation Paradigm(https://arxiv.org/abs/2602.17682)
• Keywords: diffusion, transformer, generative
• Abstract: Consistency-based generative models like Shortcut and MeanFlow achieve impressive results via a target-aware design for solving the Probability Flow ODE (PF-ODE). Typically, such methods introduce a target time $r$ alongside the current time $t$ to modulate outputs between a local multi-step derivative ($r = t$) and a global few-step integral ($r = 0$). However, the conventional "one input, one output" paradigm enforces a partition of the training budget, often allocating a significant portion (e.g., 75% in MeanFlow) solely to the multi-step objective for stability. This separation forces a trade-off: allocating sufficient samples to the multi-step objective leaves the few-step generation undertrained, which harms convergence and limits scalability. To this end, we propose Duality Models (DuMo) via a "one input, dual output" paradigm. Using a shared backbone with dual heads, DuMo simultaneously predicts velocity $v_t$ and flow-map $u_t$ from a single input $x_t$. This applies geometric constraints from the multi-step objective to every sample, bounding the few-step estimation without separating training objectives, thereby significantly improving stability and efficiency. On ImageNet 256 $\times$ 256, a 679M Diffusion Transformer with SD-VAE achieves a state-of-the-art (SOTA) FID of 1.79 in just 2 steps. Code is available at: this https URL

Title: Probabilistic NDVI Forecasting from Sparse Satellite Time Series and Weather Covariates

• Authors: Irene Iele, Giulia Romoli, Daniele Molino, Elena Mulero Ayllón, Filippo Ruffini, Paolo Soda, Matteo Tortora
• Subjects: cs.LG, cs.CV, stat.ML
• Abstract URL: https://arxiv.org/abs/2602.17683
• Pdf URL: https://arxiv.org/pdf/2602.17683
• Copy Paste: [[2602.17683]] Probabilistic NDVI Forecasting from Sparse Satellite Time Series and Weather Covariates(https://arxiv.org/abs/2602.17683)
• Keywords: transformer
• Abstract: Accurate short-term forecasting of vegetation dynamics is a key enabler for data-driven decision support in precision agriculture. Normalized Difference Vegetation Index (NDVI) forecasting from satellite observations, however, remains challenging due to sparse and irregular sampling caused by cloud coverage, as well as the heterogeneous climatic conditions under which crops evolve. In this work, we propose a probabilistic forecasting framework specifically designed for field-level NDVI prediction under clear-sky acquisition constraints. The method leverages a transformer-based architecture that explicitly separates the modeling of historical vegetation dynamics from future exogenous information, integrating historical NDVI observations with both historical and future meteorological covariates. To address irregular revisit patterns and horizon-dependent uncertainty, we introduce a temporal-distance weighted quantile loss that aligns the training objective with the effective forecasting horizon. In addition, we incorporate cumulative and extreme-weather feature engineering to better capture delayed meteorological effects relevant to vegetation response. Extensive experiments on European satellite data demonstrate that the proposed approach consistently outperforms a diverse set of statistical, deep learning, and recent time series baselines across both point-wise and probabilistic evaluation metrics. Ablation studies further highlight the central role of target history, while showing that meteorological covariates provide complementary gains when jointly exploited. The code is available at this https URL.

Title: CodeScaler: Scaling Code LLM Training and Test-Time Inference via Execution-Free Reward Models

• Authors: Xiao Zhu, Xinyu Zhou, Boyu Zhu, Hanxu Hu, Mingzhe Du, Haotian Zhang, Huiming Wang, Zhijiang Guo
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17684
• Pdf URL: https://arxiv.org/pdf/2602.17684
• Copy Paste: [[2602.17684]] CodeScaler: Scaling Code LLM Training and Test-Time Inference via Execution-Free Reward Models(https://arxiv.org/abs/2602.17684)
• Keywords: robust, extraction, large language model
• Abstract: Reinforcement Learning from Verifiable Rewards (RLVR) has driven recent progress in code large language models by leveraging execution-based feedback from unit tests, but its scalability is fundamentally constrained by the availability and reliability of high-quality test cases. We propose CodeScaler, an execution-free reward model designed to scale both reinforcement learning training and test-time inference for code generation. CodeScaler is trained on carefully curated preference data derived from verified code problems and incorporates syntax-aware code extraction and validity-preserving reward shaping to ensure stable and robust optimization. Across five coding benchmarks, CodeScaler improves Qwen3-8B-Base by an average of +11.72 points, outperforming binary execution-based RL by +1.82 points, and enables scalable reinforcement learning on synthetic datasets without any test cases. At inference time, CodeScaler serves as an effective test-time scaling method, achieving performance comparable to unit test approaches while providing a 10-fold reduction in latency. Moreover, CodeScaler surpasses existing reward models on RM-Bench not only in the code domain (+3.3 points), but also in general and reasoning domains (+2.7 points on average).

Title: Curriculum Learning for Efficient Chain-of-Thought Distillation via Structure-Aware Masking and GRPO

• Authors: Bowen Yu, Maolin Wang, Sheng Zhang, Binhao Wang, Yi Wen, Jingtong Gao, Bowen Liu, Zimo Zhao, Wanyu Wang, Xiangyu Zhao
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17686
• Pdf URL: https://arxiv.org/pdf/2602.17686
• Copy Paste: [[2602.17686]] Curriculum Learning for Efficient Chain-of-Thought Distillation via Structure-Aware Masking and GRPO(https://arxiv.org/abs/2602.17686)
• Keywords: interpretability, large language model
• Abstract: Distilling Chain-of-Thought (CoT) reasoning from large language models into compact student models presents a fundamental challenge: teacher rationales are often too verbose for smaller models to faithfully reproduce. Existing approaches either compress reasoning into single-step, losing the interpretability that makes CoT valuable. We present a three-stage curriculum learning framework that addresses this capacity mismatch through progressive skill acquisition. First, we establish structural understanding via masked shuffled reconstruction. Second, we apply Group Relative Policy Optimization (GRPO) on masked completion tasks, enabling the model to discover its own balance between accuracy and brevity. Third, we identify persistent failure cases and guide the student to internalize teacher knowledge through targeted rewriting, again optimized with GRPO. Experiments on GSM8K demonstrate that our approach enables Qwen2.5-3B-Base to achieve an 11.29 percent accuracy improvement while reducing output length by 27.4 percent, surpassing both instruction-tuned variants and prior distillation methods.

Title: AnCoder: Anchored Code Generation via Discrete Diffusion Models

• Authors: Anton Xue, Litu Rout, Constantine Caramanis, Sanjay Shakkottai
• Subjects: cs.LG, cs.PL
• Abstract URL: https://arxiv.org/abs/2602.17688
• Pdf URL: https://arxiv.org/pdf/2602.17688
• Copy Paste: [[2602.17688]] AnCoder: Anchored Code Generation via Discrete Diffusion Models(https://arxiv.org/abs/2602.17688)
• Keywords: diffusion
• Abstract: Diffusion language models offer a compelling alternative to autoregressive code generation, enabling global planning and iterative refinement of complex program logic. However, existing approaches fail to respect the rigid structure of programming languages and, as a result, often produce broken programs that fail to execute. To address this, we introduce AnchorTree, a framework that explicitly anchors the diffusion process using structured, hierarchical priors native to code. Specifically, AnchorTree uses the abstract syntax tree to prioritize resolving syntactically and semantically salient tokens, such as keywords (e.g., if, while) and identifiers (e.g., variable names), thereby establishing a structural scaffold that guides the remaining generation. We validate this framework via AnCoder, a family of models showing that structurally anchored diffusion offers a parameter-efficient path to high-quality code generation.

Title: Robust Pre-Training of Medical Vision-and-Language Models with Domain-Invariant Multi-Modal Masked Reconstruction

• Authors: Melika Filvantorkaman, Mohsen Piri
• Subjects: cs.LG, cs.AI, cs.CL, cs.CV
• Abstract URL: https://arxiv.org/abs/2602.17689
• Pdf URL: https://arxiv.org/pdf/2602.17689
• Copy Paste: [[2602.17689]] Robust Pre-Training of Medical Vision-and-Language Models with Domain-Invariant Multi-Modal Masked Reconstruction(https://arxiv.org/abs/2602.17689)
• Keywords: robust
• Abstract: Medical vision-language models show strong potential for joint reasoning over medical images and clinical text, but their performance often degrades under domain shift caused by variations in imaging devices, acquisition protocols, and reporting styles. Existing multi-modal pre-training methods largely overlook robustness, treating it as a downstream adaptation problem. In this work, we propose Robust Multi-Modal Masked Reconstruction (Robust-MMR), a self-supervised pre-training framework that explicitly incorporates robustness objectives into masked vision-language learning. Robust-MMR integrates asymmetric perturbation-aware masking, domain-consistency regularization, and modality-resilience constraints to encourage domain-invariant representations. We evaluate Robust-MMR on multiple medical vision-language benchmarks, including medical visual question answering (VQA-RAD, SLAKE, VQA-2019), cross-domain image-text classification (MELINDA), and robust image-caption retrieval (ROCO). Robust-MMR achieves 78.9% cross-domain accuracy on VQA-RAD, outperforming the strongest baseline by 3.8 percentage points, and reaches 74.6% and 77.0% accuracy on SLAKE and VQA-2019, respectively. Under perturbed evaluation, Robust-MMR improves VQA-RAD accuracy from 69.1% to 75.6%. For image-text classification, cross-domain MELINDA accuracy increases from 70.3% to 75.2%, while retrieval experiments show a reduction in mean rank degradation from over 16 to 4.1 under perturbation. Qualitative results further demonstrate improved clinical reasoning for disease detection and structural abnormality assessment. These findings show that explicitly modeling robustness during pre-training leads to more reliable and transferable medical vision-language representations for real-world deployment.

Title: Tethered Reasoning: Decoupling Entropy from Hallucination in Quantized LLMs via Manifold Steering

• Authors: Craig Atkinson
• Subjects: cs.LG, cs.CL
• Abstract URL: https://arxiv.org/abs/2602.17691
• Pdf URL: https://arxiv.org/pdf/2602.17691
• Copy Paste: [[2602.17691]] Tethered Reasoning: Decoupling Entropy from Hallucination in Quantized LLMs via Manifold Steering(https://arxiv.org/abs/2602.17691)
• Keywords: transformer
• Abstract: Quantized language models face a fundamental dilemma: low sampling temperatures yield repetitive, mode-collapsed outputs, while high temperatures (T > 2.0) cause trajectory divergence and semantic incoherence. We present HELIX, a geometric framework that decouples output entropy from hallucination by tethering hidden-state trajectories to a pre-computed truthfulness manifold. HELIX computes a Unified Truth Score (UTS) combining token-level semantic entropy with Mahalanobis distance from the manifold. When UTS indicates trajectory divergence, graduated steering vectors redirect activations toward structurally coherent regions while affecting only 0.2-2.5% of tokens. On 4-bit quantized Granite 4.0 H Small (32B/9B active, hybrid Mamba-Transformer): GSM8K maintains 88.84% accuracy at T = 3.0 (2.81pp degradation from T = 0.5); MMLU maintains 72.49% across 14,042 questions (1.24pp degradation). This demonstrates that high-temperature hallucination is primarily trajectory divergence rather than semantic collapse. Notably, steering the sparse Transformer attention layers (~10% of layers) is sufficient to correct drift in the Mamba-2 state-space formulation. Geometric tethering reveals a previously-masked High-Entropy Creative Reservoir. At T > 2.0, steered outputs exhibit 5-20% idea duplication versus 70-80% at conservative settings. Cross-architecture validation (Qwen3-30B-A3B MOE) confirms this phenomenon is architecture-independent, with 46.7% higher unique concept generation. HELIX acts as a syntax tether, enabling exploration of semantic diversity without violating the logical backbone required for valid output. This enables Multi-Temperature Synthesis, generating 200% more unique concepts than single-temperature inference.

Title: AsynDBT: Asynchronous Distributed Bilevel Tuning for efficient In-Context Learning with Large Language Models

• Authors: Hui Ma, Shaoyu Dou, Ya Liu, Fei Xing, Li Feng, Feng Pi
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17694
• Pdf URL: https://arxiv.org/pdf/2602.17694
• Copy Paste: [[2602.17694]] AsynDBT: Asynchronous Distributed Bilevel Tuning for efficient In-Context Learning with Large Language Models(https://arxiv.org/abs/2602.17694)
• Keywords: privacy, protect, federate, large language model
• Abstract: With the rapid development of large language models (LLMs), an increasing number of applications leverage cloud-based LLM APIs to reduce usage costs. However, since cloud-based models' parameters and gradients are agnostic, users have to manually or use heuristic algorithms to adjust prompts for intervening LLM outputs, which requiring costly optimization procedures. In-context learning (ICL) has recently emerged as a promising paradigm that enables LLMs to adapt to new tasks using examples provided within the input, eliminating the need for parameter updates. Nevertheless, the advancement of ICL is often hindered by the lack of high-quality data, which is often sensitive and different to share. Federated learning (FL) offers a potential solution by enabling collaborative training of distributed LLMs while preserving data privacy. Despite this issues, previous FL approaches that incorporate ICL have struggled with severe straggler problems and challenges associated with heterogeneous non-identically data. To address these problems, we propose an asynchronous distributed bilevel tuning (AsynDBT) algorithm that optimizes both in-context learning samples and prompt fragments based on the feedback from the LLM, thereby enhancing downstream task performance. Benefiting from its distributed architecture, AsynDBT provides privacy protection and adaptability to heterogeneous computing environments. Furthermore, we present a theoretical analysis establishing the convergence guarantees of the proposed algorithm. Extensive experiments conducted on multiple benchmark datasets demonstrate the effectiveness and efficiency of AsynDBT.

Title: EXACT: Explicit Attribute-Guided Decoding-Time Personalization

• Authors: Xin Yu, Hanwen Xing, Lingzhou Xue
• Subjects: cs.LG, cs.AI, cs.IR
• Abstract URL: https://arxiv.org/abs/2602.17695
• Pdf URL: https://arxiv.org/pdf/2602.17695
• Copy Paste: [[2602.17695]] EXACT: Explicit Attribute-Guided Decoding-Time Personalization(https://arxiv.org/abs/2602.17695)
• Keywords: large language model
• Abstract: Achieving personalized alignment requires adapting large language models to each user's evolving context. While decoding-time personalization offers a scalable alternative to training-time methods, existing methods largely rely on implicit, less interpretable preference representations and impose a rigid, context-agnostic user representation, failing to account for how preferences shift across prompts. We introduce EXACT, a new decoding-time personalization that aligns generation with limited pairwise preference feedback using a predefined set of interpretable attributes. EXACT first identifies user-specific attribute subsets by maximizing the likelihood of preferred responses in the offline stage. Then, for online inference, EXACT retrieves the most semantically relevant attributes for an incoming prompt and injects them into the context to steer generation. We establish theoretical approximation guarantees for the proposed algorithm under mild assumptions, and provably show that our similarity-based retrieval mechanism effectively mitigates contextual preference shifts, adapting to disparate tasks without pooling conflicting preferences. Extensive experiments on human-annotated preference datasets demonstrate that EXACT consistently outperforms strong baselines, including preference modeling accuracy and personalized generation quality.

Title: Can LLM Safety Be Ensured by Constraining Parameter Regions?

• Authors: Zongmin Li, Jian Su, Farah Benamara, Aixin Sun
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17696
• Pdf URL: https://arxiv.org/pdf/2602.17696
• Copy Paste: [[2602.17696]] Can LLM Safety Be Ensured by Constraining Parameter Regions?(https://arxiv.org/abs/2602.17696)
• Keywords: transformer, large language model
• Abstract: Large language models (LLMs) are often assumed to contain ``safety regions'' -- parameter subsets whose modification directly influences safety behaviors. We conduct a systematic evaluation of four safety region identification methods spanning different parameter granularities, from individual weights to entire Transformer layers, across four families of backbone LLMs with varying sizes. Using ten safety identification datasets, we find that the identified safety regions exhibit only low to moderate overlap, as measured by IoU. The overlap drops significantly when the safety regions are further refined using utility datasets (\ie non-harmful queries). These results suggest that current techniques fail to reliably identify a stable, dataset-agnostic safety region.

Title: Pimp My LLM: Leveraging Variability Modeling to Tune Inference Hyperparameters

• Authors: Nada Zine, Clément Quinton, Romain Rouvoy
• Subjects: cs.LG, cs.SE
• Abstract URL: https://arxiv.org/abs/2602.17697
• Pdf URL: https://arxiv.org/pdf/2602.17697
• Copy Paste: [[2602.17697]] Pimp My LLM: Leveraging Variability Modeling to Tune Inference Hyperparameters(https://arxiv.org/abs/2602.17697)
• Keywords: transformer, large language model
• Abstract: Large Language Models (LLMs) are being increasingly used across a wide range of tasks. However, their substantial computational demands raise concerns about the energy efficiency and sustainability of both training and inference. Inference, in particular, dominates total compute usage, making its optimization crucial. Recent research has explored optimization techniques and analyzed how configuration choices influence energy consumption. Yet, the vast configuration space of inference servers makes exhaustive empirical evaluation infeasible due to combinatorial explosion. In this paper, we introduce a new perspective on this problem by treating LLMs as configurable systems and applying variability management techniques to systematically analyze inference-time configuration choices. We evaluate our approach on the Hugging Face Transformers library by representing generation hyperparameters and their constraints using a feature-based variability model, sampling representative configurations, measuring their energy consumption, latency, accuracy, and learning predictive models from the collected data. Our results show that variability modeling effectively manages the complexity of LLM inference configurations. It enables systematic analysis of hyperparameters effects and interactions, reveals trade-offs, and supports accurate prediction of inference behavior from a limited number of measurements. Overall, this work opens a new research direction that bridges software engineering and machine learning by leveraging variability modeling for the efficient and sustainable configuration of LLMs.

Title: ScaleBITS: Scalable Bitwidth Search for Hardware-Aligned Mixed-Precision LLMs

• Authors: Xinlin Li, Timothy Chou, Josh Fromm, Zichang Liu, Yunjie Pan, Christina Fragouli
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17698
• Pdf URL: https://arxiv.org/pdf/2602.17698
• Copy Paste: [[2602.17698]] ScaleBITS: Scalable Bitwidth Search for Hardware-Aligned Mixed-Precision LLMs(https://arxiv.org/abs/2602.17698)
• Keywords: large language model
• Abstract: Post-training weight quantization is crucial for reducing the memory and inference cost of large language models (LLMs), yet pushing the average precision below 4 bits remains challenging due to highly non-uniform weight sensitivity and the lack of principled precision allocation. Existing solutions use irregular fine-grained mixed-precision with high runtime overhead or rely on heuristics or highly constrained precision allocation strategies. In this work, we propose ScaleBITS, a mixed-precision quantization framework that enables automated, fine-grained bitwidth allocation under a memory budget while preserving hardware efficiency. Guided by a new sensitivity analysis, we introduce a hardware-aligned, block-wise weight partitioning scheme, powered by bi-directional channel reordering. We formulate global bitwidth allocation as a constrained optimization problem and develop a scalable approximation to the greedy algorithm, enabling end-to-end principled allocation. Experiments show that ScaleBITS significantly improves over uniform-precision quantization (up to +36%) and outperforms state-of-the-art sensitivity-aware baselines (up to +13%) in ultra-low-bit regime, without adding runtime overhead.

Title: Certified Learning under Distribution Shift: Sound Verification and Identifiable Structure

• Authors: Chandrasekhar Gokavarapu, Sudhakar Gadde, Y. Rajasekhar, S. R. Bhargava (Mathematics, Government College (Autonomous), Rajahmundry, Andhra Pradesh, India)
• Subjects: cs.LG, math.RA, stat.ML
• Abstract URL: https://arxiv.org/abs/2602.17699
• Pdf URL: https://arxiv.org/pdf/2602.17699
• Copy Paste: [[2602.17699]] Certified Learning under Distribution Shift: Sound Verification and Identifiable Structure(https://arxiv.org/abs/2602.17699)
• Keywords: interpretability
• Abstract: Proposition. Let $f$ be a predictor trained on a distribution $P$ and evaluated on a shifted distribution $Q$. Under verifiable regularity and complexity constraints, the excess risk under shift admits an explicit upper bound determined by a computable shift metric and model parameters. We develop a unified framework in which (i) risk under distribution shift is certified by explicit inequalities, (ii) verification of learned models is sound for nontrivial sizes, and (iii) interpretability is enforced through identifiability conditions rather than post hoc explanations. All claims are stated with explicit assumptions. Failure modes are isolated. Non-certifiable regimes are characterized.

Title: MIDAS: Mosaic Input-Specific Differentiable Architecture Search

• Authors: Konstanty Subbotko
• Subjects: cs.LG, cs.AI, cs.NE
• Abstract URL: https://arxiv.org/abs/2602.17700
• Pdf URL: https://arxiv.org/pdf/2602.17700
• Copy Paste: [[2602.17700]] MIDAS: Mosaic Input-Specific Differentiable Architecture Search(https://arxiv.org/abs/2602.17700)
• Keywords: robust
• Abstract: Differentiable Neural Architecture Search (NAS) provides efficient, gradient-based methods for automatically designing neural networks, yet its adoption remains limited in practice. We present MIDAS, a novel approach that modernizes DARTS by replacing static architecture parameters with dynamic, input-specific parameters computed via self-attention. To improve robustness, MIDAS (i) localizes the architecture selection by computing it separately for each spatial patch of the activation map, and (ii) introduces a parameter-free, topology-aware search space that models node connectivity and simplifies selecting the two incoming edges per node. We evaluate MIDAS on the DARTS, NAS-Bench-201, and RDARTS search spaces. In DARTS, it reaches 97.42% top-1 on CIFAR-10 and 83.38% on CIFAR-100. In NAS-Bench-201, it consistently finds globally optimal architectures. In RDARTS, it sets the state of the art on two of four search spaces on CIFAR-10. We further analyze why MIDAS works, showing that patchwise attention improves discrimination among candidate operations, and the resulting input-specific parameter distributions are class-aware and predominantly unimodal, providing reliable guidance for decoding.

Title: Parallel Complex Diffusion for Scalable Time Series Generation

• Authors: Rongyao Cai, Yuxi Wan, Kexin Zhang, Ming Jin, Zhiqiang Ge, Qingsong Wen, Yong Liu
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17706
• Pdf URL: https://arxiv.org/pdf/2602.17706
• Copy Paste: [[2602.17706]] Parallel Complex Diffusion for Scalable Time Series Generation(https://arxiv.org/abs/2602.17706)
• Keywords: diffusion, generative
• Abstract: Modeling long-range dependencies in time series generation poses a fundamental trade-off between representational capacity and computational efficiency. Traditional temporal diffusion models suffer from local entanglement and the $\mathcal{O}(L^2)$ cost of attention mechanisms. We address these limitations by introducing PaCoDi (Parallel Complex Diffusion), a spectral-native architecture that decouples generative modeling in the frequency domain. PaCoDi fundamentally alters the problem topology: the Fourier Transform acts as a diagonalizing operator, converting locally coupled temporal signals into globally decorrelated spectral components. Theoretically, we prove the Quadrature Forward Diffusion and Conditional Reverse Factorization theorem, demonstrating that the complex diffusion process can be split into independent real and imaginary branches. We bridge the gap between this decoupled theory and data reality using a \textbf{Mean Field Theory (MFT) approximation} reinforced by an interactive correction mechanism. Furthermore, we generalize this discrete DDPM to continuous-time Frequency SDEs, rigorously deriving the Spectral Wiener Process describe the differential spectral Brownian motion limit. Crucially, PaCoDi exploits the Hermitian Symmetry of real-valued signals to compress the sequence length by half, achieving a 50% reduction in attention FLOPs without information loss. We further derive a rigorous Heteroscedastic Loss to handle the non-isotropic noise distribution on the compressed manifold. Extensive experiments show that PaCoDi outperforms existing baselines in both generation quality and inference speed, offering a theoretically grounded and computationally efficient solution for time series modeling.

Title: Provable Adversarial Robustness in In-Context Learning

• Authors: Di Zhang
• Subjects: cs.LG, stat.ML
• Abstract URL: https://arxiv.org/abs/2602.17743
• Pdf URL: https://arxiv.org/pdf/2602.17743
• Copy Paste: [[2602.17743]] Provable Adversarial Robustness in In-Context Learning(https://arxiv.org/abs/2602.17743)
• Keywords: robust, transformer, large language model
• Abstract: Large language models adapt to new tasks through in-context learning (ICL) without parameter updates. Current theoretical explanations for this capability assume test tasks are drawn from a distribution similar to that seen during pretraining. This assumption overlooks adversarial distribution shifts that threaten real-world reliability. To address this gap, we introduce a distributionally robust meta-learning framework that provides worst-case performance guarantees for ICL under Wasserstein-based distribution shifts. Focusing on linear self-attention Transformers, we derive a non-asymptotic bound linking adversarial perturbation strength ($\rho$), model capacity ($m$), and the number of in-context examples ($N$). The analysis reveals that model robustness scales with the square root of its capacity ($\rho_{\text{max}} \propto \sqrt{m}$), while adversarial settings impose a sample complexity penalty proportional to the square of the perturbation magnitude ($N_\rho - N_0 \propto \rho^2$). Experiments on synthetic tasks confirm these scaling laws. These findings advance the theoretical understanding of ICL's limits under adversarial conditions and suggest that model capacity serves as a fundamental resource for distributional robustness.

Title: Bayesian Optimality of In-Context Learning with Selective State Spaces

• Authors: Di Zhang, Jiaqi Xing
• Subjects: cs.LG, cs.CL, math.ST, stat.ML
• Abstract URL: https://arxiv.org/abs/2602.17744
• Pdf URL: https://arxiv.org/pdf/2602.17744
• Copy Paste: [[2602.17744]] Bayesian Optimality of In-Context Learning with Selective State Spaces(https://arxiv.org/abs/2602.17744)
• Keywords: robust, transformer
• Abstract: We propose Bayesian optimal sequential prediction as a new principle for understanding in-context learning (ICL). Unlike interpretations framing Transformers as performing implicit gradient descent, we formalize ICL as meta-learning over latent sequence tasks. For tasks governed by Linear Gaussian State Space Models (LG-SSMs), we prove a meta-trained selective SSM asymptotically implements the Bayes-optimal predictor, converging to the posterior predictive mean. We further establish a statistical separation from gradient descent, constructing tasks with temporally correlated noise where the optimal Bayesian predictor strictly outperforms any empirical risk minimization (ERM) estimator. Since Transformers can be seen as performing implicit ERM, this demonstrates selective SSMs achieve lower asymptotic risk due to superior statistical efficiency. Experiments on synthetic LG-SSM tasks and a character-level Markov benchmark confirm selective SSMs converge faster to Bayes-optimal risk, show superior sample efficiency with longer contexts in structured-noise settings, and track latent states more robustly than linear Transformers. This reframes ICL from "implicit optimization" to "optimal inference," explaining the efficiency of selective SSMs and offering a principled basis for architecture design.

Title: KPM-Bench: A Kinematic Parsing Motion Benchmark for Fine-grained Motion-centric Video Understanding

• Authors: Boda Lin, Yongjie Zhu, Xiaocheng Gong, Wenyu Qin, Meng Wang
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.17768
• Pdf URL: https://arxiv.org/pdf/2602.17768
• Copy Paste: [[2602.17768]] KPM-Bench: A Kinematic Parsing Motion Benchmark for Fine-grained Motion-centric Video Understanding(https://arxiv.org/abs/2602.17768)
• Keywords: extraction
• Abstract: Despite recent advancements, video captioning models still face significant limitations in accurately describing fine-grained motion details and suffer from severe hallucination issues. These challenges become particularly prominent when generating captions for motion-centric videos, where precise depiction of intricate movements and limb dynamics is crucial yet often neglected. To alleviate this gap, we introduce an automated annotation pipeline that integrates kinematic-based motion computation with linguistic parsing, enabling detailed decomposition and description of complex human motions. Based on this pipeline, we construct and release the Kinematic Parsing Motion Benchmark (KPM-Bench), a novel open-source dataset designed to facilitate fine-grained motion understanding. KPM-Bench consists of (i) fine-grained video-caption pairs that comprehensively illustrate limb-level dynamics in complex actions, (ii) diverse and challenging question-answer pairs focusing specifically on motion understanding, and (iii) a meticulously curated evaluation set specifically designed to assess hallucination phenomena associated with motion descriptions. Furthermore, to address hallucination issues systematically, we propose the linguistically grounded Motion Parsing and Extraction (MoPE) algorithm, capable of accurately extracting motion-specific attributes directly from textual captions. Leveraging MoPE, we introduce a precise hallucination evaluation metric that functions independently of large-scale vision-language or language-only models. By integrating MoPE into the GRPO post-training framework, we effectively mitigate hallucination problems, significantly improving the reliability of motion-centric video captioning models.

Title: Asking Forever: Universal Activations Behind Turn Amplification in Conversational LLMs

• Authors: Zachary Coalson, Bo Fang, Sanghyun Hong
• Subjects: cs.LG, cs.CR
• Abstract URL: https://arxiv.org/abs/2602.17778
• Pdf URL: https://arxiv.org/pdf/2602.17778
• Copy Paste: [[2602.17778]] Asking Forever: Universal Activations Behind Turn Amplification in Conversational LLMs(https://arxiv.org/abs/2602.17778)
• Keywords: protect, defense, attack
• Abstract: Multi-turn interaction length is a dominant factor in the operational costs of conversational LLMs. In this work, we present a new failure mode in conversational LLMs: turn amplification, in which a model consistently prolongs multi-turn interactions without completing the underlying task. We show that an adversary can systematically exploit clarification-seeking behavior$-$commonly encouraged in multi-turn conversation settings$-$to scalably prolong interactions. Moving beyond prompt-level behaviors, we take a mechanistic perspective and identify a query-independent, universal activation subspace associated with clarification-seeking responses. Unlike prior cost-amplification attacks that rely on per-turn prompt optimization, our attack arises from conversational dynamics and persists across prompts and tasks. We show that this mechanism provides a scalable pathway to induce turn amplification: both supply-chain attacks via fine-tuning and runtime attacks through low-level parameter corruptions consistently shift models toward abstract, clarification-seeking behavior across prompts. Across multiple instruction-tuned LLMs and benchmarks, our attack substantially increases turn count while remaining compliant. We also show that existing defenses offer limited protection against this emerging class of failures.

Title: Enabling Training-Free Text-Based Remote Sensing Segmentation

• Authors: Jose Sosa, Danila Rukhovich, Anis Kacem, Djamila Aouada
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.17799
• Pdf URL: https://arxiv.org/pdf/2602.17799
• Copy Paste: [[2602.17799]] Enabling Training-Free Text-Based Remote Sensing Segmentation(https://arxiv.org/abs/2602.17799)
• Keywords: generative, segmentation
• Abstract: Recent advances in Vision Language Models (VLMs) and Vision Foundation Models (VFMs) have opened new opportunities for zero-shot text-guided segmentation of remote sensing imagery. However, most existing approaches still rely on additional trainable components, limiting their generalisation and practical applicability. In this work, we investigate to what extent text-based remote sensing segmentation can be achieved without additional training, by relying solely on existing foundation models. We propose a simple yet effective approach that integrates contrastive and generative VLMs with the Segment Anything Model (SAM), enabling a fully training-free or lightweight LoRA-tuned pipeline. Our contrastive approach employs CLIP as mask selector for SAM's grid-based proposals, achieving state-of-the-art open-vocabulary semantic segmentation (OVSS) in a completely zero-shot setting. In parallel, our generative approach enables reasoning and referring segmentation by generating click prompts for SAM using GPT-5 in a zero-shot setting and a LoRA-tuned Qwen-VL model, with the latter yielding the best results. Extensive experiments across 19 remote sensing benchmarks, including open-vocabulary, referring, and reasoning-based tasks, demonstrate the strong capabilities of our approach. Code will be released at this https URL.

Title: Exploiting Liquidity Exhaustion Attacks in Intent-Based Cross-Chain Bridges

• Authors: André Augusto, Christof Ferreira Torres, André Vasconcelos, Miguel Correia
• Subjects: cs.CR
• Abstract URL: https://arxiv.org/abs/2602.17805
• Pdf URL: https://arxiv.org/pdf/2602.17805
• Copy Paste: [[2602.17805]] Exploiting Liquidity Exhaustion Attacks in Intent-Based Cross-Chain Bridges(https://arxiv.org/abs/2602.17805)
• Keywords: secure, attack, robust
• Abstract: Intent-based cross-chain bridges have emerged as an alternative to traditional interoperability protocols by allowing off-chain entities (\emph{solvers}) to immediately fulfill users' orders by fronting their own liquidity. While improving user experience, this approach introduces new systemic risks, such as solver liquidity concentration and delayed settlement. In this paper, we propose a new class of attacks called \emph{liquidity exhaustion attacks} and a replay-based parameterized attack simulation framework. We analyze 3.5 million cross-chain intents that moved \$9.24B worth of tokens between June and November 2025 across three major protocols (Mayan Swift, Across, and deBridge), spanning nine blockchains. For rational attackers, our results show that protocols with higher solver profitability, such as deBridge, are vulnerable under current parameters: 210 historical attack instances yield a mean net profit of \$286.14, with 80.5\% of attacks profitable. In contrast, Across remains robust in all tested configurations due to low solver margins and very high liquidity, while Mayan Swift is generally secure but becomes vulnerable under stress-test conditions. Under byzantine attacks, we show that it is possible to suppress availability across all protocols, causing dozens of failed intents and solver profit losses of up to \$978 roughly every 16 minutes. Finally, we propose an optimized attack strategy that exploits patterns in the data to reduce attack costs by up to 90.5\% compared to the baseline, lowering the barrier to liquidity exhaustion attacks.

Title: VidEoMT: Your ViT is Secretly Also a Video Segmentation Model

• Authors: Narges Norouzi, Idil Esen Zulfikar, Niccol`o Cavagnero, Tommie Kerssies, Bastian Leibe, Gijs Dubbelman, Daan de Geus
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.17807
• Pdf URL: https://arxiv.org/pdf/2602.17807
• Copy Paste: [[2602.17807]] VidEoMT: Your ViT is Secretly Also a Video Segmentation Model(https://arxiv.org/abs/2602.17807)
• Keywords: transformer, segmentation
• Abstract: Existing online video segmentation models typically combine a per-frame segmenter with complex specialized tracking modules. While effective, these modules introduce significant architectural complexity and computational overhead. Recent studies suggest that plain Vision Transformer (ViT) encoders, when scaled with sufficient capacity and large-scale pre-training, can conduct accurate image segmentation without requiring specialized modules. Motivated by this observation, we propose the Video Encoder-only Mask Transformer (VidEoMT), a simple encoder-only video segmentation model that eliminates the need for dedicated tracking modules. To enable temporal modeling in an encoder-only ViT, VidEoMT introduces a lightweight query propagation mechanism that carries information across frames by reusing queries from the previous frame. To balance this with adaptability to new content, it employs a query fusion strategy that combines the propagated queries with a set of temporally-agnostic learned queries. As a result, VidEoMT attains the benefits of a tracker without added complexity, achieving competitive accuracy while being 5x--10x faster, running at up to 160 FPS with a ViT-L backbone. Code: this https URL

Title: Calibrated Adaptation: Bayesian Stiefel Manifold Priors for Reliable Parameter-Efficient Fine-Tuning

• Authors: Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17809
• Pdf URL: https://arxiv.org/pdf/2602.17809
• Copy Paste: [[2602.17809]] Calibrated Adaptation: Bayesian Stiefel Manifold Priors for Reliable Parameter-Efficient Fine-Tuning(https://arxiv.org/abs/2602.17809)
• Keywords: large language model
• Abstract: Parameter-efficient fine-tuning methods such as LoRA enable practical adaptation of large language models but provide no principled uncertainty estimates, leading to poorly calibrated predictions and unreliable behavior under domain shift. We introduce Stiefel-Bayes Adapters (SBA), a Bayesian PEFT framework that places a Matrix Langevin prior over orthonormal adapter factors on the Stiefel manifold $\St$ and performs approximate posterior inference via tangent space Laplace approximation with geodesic retraction. Unlike Gaussian priors in flat space projected onto orthogonality constraints, our prior on the manifold naturally encodes the inductive bias that adapter subspaces should be well conditioned and orthogonal, while the posterior provides calibrated predictive uncertainty without recalibration. We prove formally that the tangent space approximation strictly avoids the structural variance inflation inherent in projecting from ambient space, establishing a rigorous theoretical advantage for intrinsic manifold inference. Across GLUE and SuperGLUE benchmarks on RoBERTa-large, LLaMA-2-7B, LLaMA-2-13B, Mistral-7B, and Qwen2.5-7B, domain shift evaluations, selective prediction protocols, and an abstractive summarization task, SBA achieves task performance comparable to LoRA and DoRA while reducing Expected Calibration Error by 18 to 34\% over deterministic baselines, improving selective prediction AUROC by 12 to 25\% under domain shift, and outperforming deep ensembles of five LoRA models on OOD detection at a fraction of the parameter cost. Our results demonstrate that where you place uncertainty, on the right geometric structure, matters more than simply adding any Bayesian treatment to adapters.

Title: VQPP: Video Query Performance Prediction Benchmark

• Authors: Adrian Catalin Lutu, Eduard Poesina, Radu Tudor Ionescu
• Subjects: cs.CV, cs.IR, cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17814
• Pdf URL: https://arxiv.org/pdf/2602.17814
• Copy Paste: [[2602.17814]] VQPP: Video Query Performance Prediction Benchmark(https://arxiv.org/abs/2602.17814)
• Keywords: large language model
• Abstract: Query performance prediction (QPP) is an important and actively studied information retrieval task, having various applications, such as query reformulation, query expansion, and retrieval system selection, among many others. The task has been primarily studied in the context of text and image retrieval, whereas QPP for content-based video retrieval (CBVR) remains largely underexplored. To this end, we propose the first benchmark for video query performance prediction (VQPP), comprising two text-to-video retrieval datasets and two CBVR systems, respectively. VQPP contains a total of 56K text queries and 51K videos, and comes with official training, validation and test splits, fostering direct comparisons and reproducible results. We explore multiple pre-retrieval and post-retrieval performance predictors, creating a representative benchmark for future exploration of QPP in the video domain. Our results show that pre-retrieval predictors obtain competitive performance, enabling applications before performing the retrieval step. We also demonstrate the applicability of VQPP by employing the best performing pre-retrieval predictor as reward model for training a large language model (LLM) on the query reformulation task via direct preference optimization (DPO). We release our benchmark and code at this https URL.

Title: Neural Synchrony Between Socially Interacting Language Models

• Authors: Zhining Zhang, Wentao Zhu, Chi Han, Yizhou Wang, Heng Ji
• Subjects: cs.CL
• Abstract URL: https://arxiv.org/abs/2602.17815
• Pdf URL: https://arxiv.org/pdf/2602.17815
• Copy Paste: [[2602.17815]] Neural Synchrony Between Socially Interacting Language Models(https://arxiv.org/abs/2602.17815)
• Keywords: large language model
• Abstract: Neuroscience has uncovered a fundamental mechanism of our social nature: human brain activity becomes synchronized with others in many social contexts involving interaction. Traditionally, social minds have been regarded as an exclusive property of living beings. Although large language models (LLMs) are widely accepted as powerful approximations of human behavior, with multi-LLM system being extensively explored to enhance their capabilities, it remains controversial whether they can be meaningfully compared to human social minds. In this work, we explore neural synchrony between socially interacting LLMs as an empirical evidence for this debate. Specifically, we introduce neural synchrony during social simulations as a novel proxy for analyzing the sociality of LLMs at the representational level. Through carefully designed experiments, we demonstrate that it reliably reflects both social engagement and temporal alignment in their interactions. Our findings indicate that neural synchrony between LLMs is strongly correlated with their social performance, highlighting an important link between neural synchrony and the social behaviors of LLMs. Our work offers a new perspective to examine the "social minds" of LLMs, highlighting surprising parallels in the internal dynamics that underlie human and LLM social interaction.

Title: Avoid What You Know: Divergent Trajectory Balance for GFlowNets

• Authors: Pedro Dall'Antonia, Tiago da Silva, Daniel Csillag, Salem Lahlou, Diego Mesquita
• Subjects: cs.LG, stat.ML
• Abstract URL: https://arxiv.org/abs/2602.17827
• Pdf URL: https://arxiv.org/pdf/2602.17827
• Copy Paste: [[2602.17827]] Avoid What You Know: Divergent Trajectory Balance for GFlowNets(https://arxiv.org/abs/2602.17827)
• Keywords: generative
• Abstract: Generative Flow Networks (GFlowNets) are a flexible family of amortized samplers trained to generate discrete and compositional objects with probability proportional to a reward function. However, learning efficiency is constrained by the model's ability to rapidly explore diverse high-probability regions during training. To mitigate this issue, recent works have focused on incentivizing the exploration of unvisited and valuable states via curiosity-driven search and self-supervised random network distillation, which tend to waste samples on already well-approximated regions of the state space. In this context, we propose Adaptive Complementary Exploration (ACE), a principled algorithm for the effective exploration of novel and high-probability regions when learning GFlowNets. To achieve this, ACE introduces an exploration GFlowNet explicitly trained to search for high-reward states in regions underexplored by the canonical GFlowNet, which learns to sample from the target distribution. Through extensive experiments, we show that ACE significantly improves upon prior work in terms of approximation accuracy to the target distribution and discovery rate of diverse high-reward states.

Title: Causality by Abstraction: Symbolic Rule Learning in Multivariate Timeseries with Large Language Models

• Authors: Preetom Biswas, Giulia Pedrielli, K. Selçuk Candan
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17829
• Pdf URL: https://arxiv.org/pdf/2602.17829
• Copy Paste: [[2602.17829]] Causality by Abstraction: Symbolic Rule Learning in Multivariate Timeseries with Large Language Models(https://arxiv.org/abs/2602.17829)
• Keywords: large language model
• Abstract: Inferring causal relations in timeseries data with delayed effects is a fundamental challenge, especially when the underlying system exhibits complex dynamics that cannot be captured by simple functional mappings. Traditional approaches often fail to produce generalized and interpretable explanations, as multiple distinct input trajectories may yield nearly indistinguishable outputs. In this work, we present ruleXplain, a framework that leverages Large Language Models (LLMs) to extract formal explanations for input-output relations in simulation-driven dynamical systems. Our method introduces a constrained symbolic rule language with temporal operators and delay semantics, enabling LLMs to generate verifiable causal rules through structured prompting. ruleXplain relies on the availability of a principled model (e.g., a simulator) that maps multivariate input time series to output time series. Within ruleXplain, the simulator is used to generate diverse counterfactual input trajectories that yield similar target output, serving as candidate explanations. Such counterfactual inputs are clustered and provided as context to the LLM, which is tasked with the generation of symbolic rules encoding the joint temporal trends responsible for the patterns observable in the output times series. A closed-loop refinement process ensures rule consistency and semantic validity. We validate the framework using the PySIRTEM epidemic simulator, mapping testing rate inputs to daily infection counts; and the EnergyPlus building energy simulator, observing temperature and solar irradiance inputs to electricity needs. For validation, we perform three classes of experiments: (1) the efficacy of the ruleset through input reconstruction; (2) ablation studies evaluating the causal encoding of the ruleset; and (3) generalization tests of the extracted rules across unseen output trends with varying phase dynamics.

Title: MePoly: Max Entropy Polynomial Policy Optimization

• Authors: Hang Liu, Sangli Teng, Maani Ghaffari
• Subjects: cs.LG, cs.RO
• Abstract URL: https://arxiv.org/abs/2602.17832
• Pdf URL: https://arxiv.org/pdf/2602.17832
• Copy Paste: [[2602.17832]] MePoly: Max Entropy Polynomial Policy Optimization(https://arxiv.org/abs/2602.17832)
• Keywords: diffusion
• Abstract: Stochastic Optimal Control provides a unified mathematical framework for solving complex decision-making problems, encompassing paradigms such as maximum entropy reinforcement learning(RL) and imitation learning(IL). However, conventional parametric policies often struggle to represent the multi-modality of the solutions. Though diffusion-based policies are aimed at recovering the multi-modality, they lack an explicit probability density, which complicates policy-gradient optimization. To bridge this gap, we propose MePoly, a novel policy parameterization based on polynomial energy-based models. MePoly provides an explicit, tractable probability density, enabling exact entropy maximization. Theoretically, we ground our method in the classical moment problem, leveraging the universal approximation capabilities for arbitrary distributions. Empirically, we demonstrate that MePoly effectively captures complex non-convex manifolds and outperforms baselines in performance across diverse benchmarks.

Title: Influence-Preserving Proxies for Gradient-Based Data Selection in LLM Fine-tuning

• Authors: Sirui Chen, Yunzhe Qi, Mengting Ai, Yifan Sun, Ruizhong Qiu, Jiaru Zou, Jingrui He
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17835
• Pdf URL: https://arxiv.org/pdf/2602.17835
• Copy Paste: [[2602.17835]] Influence-Preserving Proxies for Gradient-Based Data Selection in LLM Fine-tuning(https://arxiv.org/abs/2602.17835)
• Keywords: large language model
• Abstract: Supervised fine-tuning (SFT) relies critically on selecting training data that most benefits a model's downstream performance. Gradient-based data selection methods such as TracIn and Influence Functions leverage influence to identify useful samples, but their computational cost scales poorly, making them impractical for multi-billion-parameter large language models (LLMs). A common alternative is to use off-the-shelf smaller models as proxies, but they remain suboptimal since their learning dynamics are unclear, their sizes cannot be flexibly adjusted, and they cannot be further aligned with the target model in terms of gradient-based influence estimation. To address these challenges, we introduce Iprox, a two-stage framework that derives influence-preserving proxies directly from the target model. It first applies a low-rank compression stage to preserve influence information of the target model, and then an aligning stage to align both model gradients and logits, thereby constructing proxies that flexibly control computational cost while retaining the target model's influence. Experimental results across diverse LLM families and evaluation tasks show that Iprox consistently outperforms off-the-shelf proxies and baseline methods. On Qwen3-4B, a 1.5B proxy constructed with Iprox achieves stronger performance than the larger 1.7B off-the-shelf proxy. Notably, on Llama3.2, Iprox achieves better performance than baselines while reducing computational cost by more than half relative to the full 3B model. These results show that Iprox provides effective influence-preserving proxies, making gradient-based data selection more scalable for LLMs.

Title: TFL: Targeted Bit-Flip Attack on Large Language Model

• Authors: Jingkai Guo, Chaitali Chakrabarti, Deliang Fan
• Subjects: cs.CR, cs.CL, cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17837
• Pdf URL: https://arxiv.org/pdf/2602.17837
• Copy Paste: [[2602.17837]] TFL: Targeted Bit-Flip Attack on Large Language Model(https://arxiv.org/abs/2602.17837)
• Keywords: security, attack, robust, steal, generative, large language model
• Abstract: Large language models (LLMs) are increasingly deployed in safety and security critical applications, raising concerns about their robustness to model parameter fault injection attacks. Recent studies have shown that bit-flip attacks (BFAs), which exploit computer main memory (i.e., DRAM) vulnerabilities to flip a small number of bits in model weights, can severely disrupt LLM behavior. However, existing BFA on LLM largely induce un-targeted failure or general performance degradation, offering limited control over manipulating specific or targeted outputs. In this paper, we present TFL, a novel targeted bit-flip attack framework that enables precise manipulation of LLM outputs for selected prompts while maintaining almost no or minor degradation on unrelated inputs. Within our TFL framework, we propose a novel keyword-focused attack loss to promote attacker-specified target tokens in generative outputs, together with an auxiliary utility score that balances attack effectiveness against collateral performance impact on benign data. We evaluate TFL on multiple LLMs (Qwen, DeepSeek, Llama) and benchmarks (DROP, GSM8K, and TriviaQA). The experiments show that TFL achieves successful targeted LLM output manipulations with less than 50 bit flips and significantly reduced effect on unrelated queries compared to prior BFA approaches. This demonstrates the effectiveness of TFL and positions it as a new class of stealthy and targeted LLM model attack.

Title: StableAML: Machine Learning for Behavioral Wallet Detection in Stablecoin Anti-Money Laundering on Ethereum

• Authors: Luciano Juvinski, Haochen Li, Alessio Brini
• Subjects: cs.CR, cs.CE
• Abstract URL: https://arxiv.org/abs/2602.17842
• Pdf URL: https://arxiv.org/pdf/2602.17842
• Copy Paste: [[2602.17842]] StableAML: Machine Learning for Behavioral Wallet Detection in Stablecoin Anti-Money Laundering on Ethereum(https://arxiv.org/abs/2602.17842)
• Keywords: robust, interpretability
• Abstract: Global illicit fund flows exceed an estimated $3.1 trillion annually, with stablecoins emerging as a preferred laundering medium due to their liquidity. While decentralized protocols increasingly adopt zero-knowledge proofs to obfuscate transaction graphs, centralized stablecoins remain critical "transparent choke points" for compliance. Leveraging this persistent visibility, this study analyzes an Ethereum dataset and uses behavioral features to develop a robust AML framework. Our findings demonstrate that domain-informed tree ensemble models achieve higher Macro-F1 score, significantly outperforming graph neural networks, which struggle with the increasing fragmentation of transaction networks. The model's interpretability goes beyond binary detection, successfully dissecting distinct typologies: it differentiates the complex, high-velocity dispersion of cybercrime syndicates from the constrained, static footprints left by sanctioned entities. This framework aligns with the industry shift toward deterministic verification, satisfying the auditability and compliance expectations under regulations such as the EU's MiCA and the U.S. GENIUS Act while minimizing unjustified asset freezes. By automating high-precision detection, we propose an approach that effectively raises the economic cost of financial misconduct without stifling innovation.

Title: Two Calm Ends and the Wild Middle: A Geometric Picture of Memorization in Diffusion Models

• Authors: Nick Dodson, Xinyu Gao, Qingsong Wang, Yusu Wang, Zhengchao Wan
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17846
• Pdf URL: https://arxiv.org/pdf/2602.17846
• Copy Paste: [[2602.17846]] Two Calm Ends and the Wild Middle: A Geometric Picture of Memorization in Diffusion Models(https://arxiv.org/abs/2602.17846)
• Keywords: privacy, diffusion
• Abstract: Diffusion models generate high-quality samples but can also memorize training data, raising serious privacy concerns. Understanding the mechanisms governing when memorization versus generalization occurs remains an active area of research. In particular, it is unclear where along the noise schedule memorization is induced, how data geometry influences it, and how phenomena at different noise scales interact. We introduce a geometric framework that partitions the noise schedule into three regimes based on the coverage properties of training data by Gaussian shells and the concentration behavior of the posterior, which we argue are two fundamental objects governing memorization and generalization in diffusion models. This perspective reveals that memorization risk is highly non-uniform across noise levels. We further identify a danger zone at medium noise levels where memorization is most pronounced. In contrast, both the small and large noise regimes resist memorization, but through fundamentally different mechanisms: small noise avoids memorization due to limited training coverage, while large noise exhibits low posterior concentration and admits a provably near linear Gaussian denoising behavior. For the medium noise regime, we identify geometric conditions through which we propose a geometry-informed targeted intervention that mitigates memorization.

Title: Dual Length Codes for Lossless Compression of BFloat16

• Authors: Aditya Agrawal, Albert Magyar, Hiteshwar Eswaraiah, Patrick Sheridan, Pradeep Janedula, Ravi Krishnan Venkatesan, Krishna Nair, Ravi Iyer
• Subjects: cs.LG, cs.IT
• Abstract URL: https://arxiv.org/abs/2602.17849
• Pdf URL: https://arxiv.org/pdf/2602.17849
• Copy Paste: [[2602.17849]] Dual Length Codes for Lossless Compression of BFloat16(https://arxiv.org/abs/2602.17849)
• Keywords: large language model
• Abstract: Training and serving Large Language Models (LLMs) relies heavily on parallelization and collective operations, which are frequently bottlenecked by network bandwidth. Lossless compression using e.g., Huffman codes can alleviate the issue, however, Huffman codes suffer from slow, bit-sequential decoding and high hardware complexity due to deep tree traversals. Universal codes e.g., Exponential-Golomb codes are faster to decode but do not exploit the symbol frequency distributions. To address these limitations, this paper introduces Dual Length Codes, a hybrid approach designed to balance compression efficiency with decoding speed. Analyzing BFloat16 tensors from the Gemma model, we observed that the top 8 most frequent symbols account for approximately 50% of the cumulative probability. These 8 symbols are assigned a short 4 bit code. The remaining 248 symbols are assigned a longer 9 bit code. The coding scheme uses a single prefix bit to distinguish between the two code lengths. The scheme uses a small Look Up Table with only 8 entries for encoding and decoding. The scheme achieves a compressibility of 18.6% in comparison to 21.3% achieved by Huffman codes, but it significantly speeds up the decoding and simplifies the hardware complexity.

Title: Neural Prior Estimation: Learning Class Priors from Latent Representations

• Authors: Masoud Yavari, Payman Moallem
• Subjects: cs.LG, cs.CV
• Abstract URL: https://arxiv.org/abs/2602.17853
• Pdf URL: https://arxiv.org/pdf/2602.17853
• Copy Paste: [[2602.17853]] Neural Prior Estimation: Learning Class Priors from Latent Representations(https://arxiv.org/abs/2602.17853)
• Keywords: segmentation
• Abstract: Class imbalance induces systematic bias in deep neural networks by imposing a skewed effective class prior. This work introduces the Neural Prior Estimator (NPE), a framework that learns feature-conditioned log-prior estimates from latent representations. NPE employs one or more Prior Estimation Modules trained jointly with the backbone via a one-way logistic loss. Under the Neural Collapse regime, NPE is analytically shown to recover the class log-prior up to an additive constant, providing a theoretically grounded adaptive signal without requiring explicit class counts or distribution-specific hyperparameters. The learned estimate is incorporated into logit adjustment, forming NPE-LA, a principled mechanism for bias-aware prediction. Experiments on long-tailed CIFAR and imbalanced semantic segmentation benchmarks (STARE, ADE20K) demonstrate consistent improvements, particularly for underrepresented classes. NPE thus offers a lightweight and theoretically justified approach to learned prior estimation and imbalance-aware prediction.

Title: JAX-Privacy: A library for differentially private machine learning

• Authors: Ryan McKenna, Galen Andrew, Borja Balle, Vadym Doroshenko, Arun Ganesh, Weiwei Kong, Alex Kurakin, Brendan McMahan, Mikhail Pravilov
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17861
• Pdf URL: https://arxiv.org/pdf/2602.17861
• Copy Paste: [[2602.17861]] JAX-Privacy: A library for differentially private machine learning(https://arxiv.org/abs/2602.17861)
• Keywords: privacy, robust
• Abstract: JAX-Privacy is a library designed to simplify the deployment of robust and performant mechanisms for differentially private machine learning. Guided by design principles of usability, flexibility, and efficiency, JAX-Privacy serves both researchers requiring deep customization and practitioners who want a more out-of-the-box experience. The library provides verified, modular primitives for critical components for all aspects of the mechanism design including batch selection, gradient clipping, noise addition, accounting, and auditing, and brings together a large body of recent research on differentially private ML.

Title: Financial time series augmentation using transformer based GAN architecture

• Authors: Andrzej Podobiński, Jarosław A. Chudziak
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17865
• Pdf URL: https://arxiv.org/pdf/2602.17865
• Copy Paste: [[2602.17865]] Financial time series augmentation using transformer based GAN architecture(https://arxiv.org/abs/2602.17865)
• Keywords: transformer, generative
• Abstract: Time-series forecasting is a critical task across many domains, from engineering to economics, where accurate predictions drive strategic decisions. However, applying advanced deep learning models in challenging, volatile domains like finance is difficult due to the inherent limitation and dynamic nature of financial time series data. This scarcity often results in sub-optimal model training and poor generalization. The fundamental challenge lies in determining how to reliably augment scarce financial time series data to enhance the predictive accuracy of deep learning forecasting models. Our main contribution is a demonstration of how Generative Adversarial Networks (GANs) can effectively serve as a data augmentation tool to overcome data scarcity in the financial domain. Specifically, we show that training a Long Short-Term Memory (LSTM) forecasting model on a dataset augmented with synthetic data generated by a transformer-based GAN (TTS-GAN) significantly improves the forecasting accuracy compared to using real data alone. We confirm these results across different financial time series (Bitcoin and S\&P500 price data) and various forecasting horizons. Furthermore, we propose a novel, time series specific quality metric that combines Dynamic Time Warping (DTW) and a modified Deep Dataset Dissimilarity Measure (DeD-iMs) to reliably monitor the training progress and evaluate the quality of the generated data. These findings provide compelling evidence for the benefits of GAN-based data augmentation in enhancing financial predictive capabilities.

Title: MantisV2: Closing the Zero-Shot Gap in Time Series Classification with Synthetic Data and Test-Time Strategies

• Authors: Vasilii Feofanov, Songkang Wen, Jianfeng Zhang, Lujia Pan, Ievgen Redko
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17868
• Pdf URL: https://arxiv.org/pdf/2602.17868
• Copy Paste: [[2602.17868]] MantisV2: Closing the Zero-Shot Gap in Time Series Classification with Synthetic Data and Test-Time Strategies(https://arxiv.org/abs/2602.17868)
• Keywords: extraction
• Abstract: Developing foundation models for time series classification is of high practical relevance, as such models can serve as universal feature extractors for diverse downstream tasks. Although early models such as Mantis have shown the promise of this approach, a substantial performance gap remained between frozen and fine-tuned encoders. In this work, we introduce methods that significantly strengthen zero-shot feature extraction for time series. First, we introduce Mantis+, a variant of Mantis pre-trained entirely on synthetic time series. Second, through controlled ablation studies, we refine the architecture and obtain MantisV2, an improved and more lightweight encoder. Third, we propose an enhanced test-time methodology that leverages intermediate-layer representations and refines output-token aggregation. In addition, we show that performance can be further improved via self-ensembling and cross-model embedding fusion. Extensive experiments on UCR, UEA, Human Activity Recognition (HAR) benchmarks, and EEG datasets show that MantisV2 and Mantis+ consistently outperform prior time series foundation models, achieving state-of-the-art zero-shot performance.

Title: Learning Compact Video Representations for Efficient Long-form Video Understanding in Large Multimodal Models

• Authors: Yuxiao Chen, Jue Wang, Zhikang Zhang, Jingru Yi, Xu Zhang, Yang Zou, Zhaowei Cai, Jianbo Yuan, Xinyu Li, Hao Yang, Davide Modolo
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.17869
• Pdf URL: https://arxiv.org/pdf/2602.17869
• Copy Paste: [[2602.17869]] Learning Compact Video Representations for Efficient Long-form Video Understanding in Large Multimodal Models(https://arxiv.org/abs/2602.17869)
• Keywords: large language model
• Abstract: With recent advancements in video backbone architectures, combined with the remarkable achievements of large language models (LLMs), the analysis of long-form videos spanning tens of minutes has become both feasible and increasingly prevalent. However, the inherently redundant nature of video sequences poses significant challenges for contemporary state-of-the-art models. These challenges stem from two primary aspects: 1) efficiently incorporating a larger number of frames within memory constraints, and 2) extracting discriminative information from the vast volume of input data. In this paper, we introduce a novel end-to-end schema for long-form video understanding, which includes an information-density-based adaptive video sampler (AVS) and an autoencoder-based spatiotemporal video compressor (SVC) integrated with a multimodal large language model (MLLM). Our proposed system offers two major advantages: it adaptively and effectively captures essential information from video sequences of varying durations, and it achieves high compression rates while preserving crucial discriminative information. The proposed framework demonstrates promising performance across various benchmarks, excelling in both long-form video understanding tasks and standard video understanding benchmarks. These results underscore the versatility and efficacy of our approach, particularly in managing the complexities of prolonged video sequences.

Title: Understanding Unreliability of Steering Vectors in Language Models: Geometric Predictors and the Limits of Linear Approximations

• Authors: Joschka Braun
• Subjects: cs.CL, cs.AI, cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17881
• Pdf URL: https://arxiv.org/pdf/2602.17881
• Copy Paste: [[2602.17881]] Understanding Unreliability of Steering Vectors in Language Models: Geometric Predictors and the Limits of Linear Approximations(https://arxiv.org/abs/2602.17881)
• Keywords: robust
• Abstract: Steering vectors are a lightweight method for controlling language model behavior by adding a learned bias to the activations at inference time. Although effective on average, steering effect sizes vary across samples and are unreliable for many target behaviors. In my thesis, I investigate why steering reliability differs across behaviors and how it is impacted by steering vector training data. First, I find that higher cosine similarity between training activation differences predicts more reliable steering. Second, I observe that behavior datasets where positive and negative activations are better separated along the steering direction are more reliably steerable. Finally, steering vectors trained on different prompt variations are directionally distinct, yet perform similarly well and exhibit correlated efficacy across datasets. My findings suggest that steering vectors are unreliable when the latent target behavior representation is not effectively approximated by the linear steering direction. Taken together, these insights offer a practical diagnostic for steering unreliability and motivate the development of more robust steering methods that explicitly account for non-linear latent behavior representations.

Title: COMBA: Cross Batch Aggregation for Learning Large Graphs with Context Gating State Space Models

• Authors: Jiajun Shen, Yufei Jin, Yi He, xingquan Zhu
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17893
• Pdf URL: https://arxiv.org/pdf/2602.17893
• Copy Paste: [[2602.17893]] COMBA: Cross Batch Aggregation for Learning Large Graphs with Context Gating State Space Models(https://arxiv.org/abs/2602.17893)
• Keywords: transformer
• Abstract: State space models (SSMs) have recently emerged for modeling long-range dependency in sequence data, with much simplified computational costs than modern alternatives, such as transformers. Advancing SMMs to graph structured data, especially for large graphs, is a significant challenge because SSMs are sequence models and the shear graph volumes make it very expensive to convert graphs as sequences for effective learning. In this paper, we propose COMBA to tackle large graph learning using state space models, with two key innovations: graph context gating and cross batch aggregation. Graph context refers to different hops of neighborhood for each node, and graph context gating allows COMBA to use such context to learn best control of neighbor aggregation. For each graph context, COMBA samples nodes as batches, and train a graph neural network (GNN), with information being aggregated cross batches, allowing COMBA to scale to large graphs. Our theoretical study asserts that cross-batch aggregation guarantees lower error than training GNN without aggregation. Experiments on benchmark networks demonstrate significant performance gains compared to baseline approaches. Code and benchmark datasets will be released for public access.

Title: Symfrog-512: High-Capacity Sponge-Based AEAD Cipher (1024-bit State)

• Authors: Victor Duarte Melo
• Subjects: cs.CR
• Abstract URL: https://arxiv.org/abs/2602.17900
• Pdf URL: https://arxiv.org/pdf/2602.17900
• Copy Paste: [[2602.17900]] Symfrog-512: High-Capacity Sponge-Based AEAD Cipher (1024-bit State)(https://arxiv.org/abs/2602.17900)
• Keywords: security
• Abstract: This submission includes a complete reference implementation together with deterministic test vectors and a reproducible benchmark suite. All source code, build instructions, and regression artifacts are publicly available in the project repository, enabling independent verification and reimplementation of the scheme. The AEAD construction is fully specified, including domain separation, rate and capacity choices, tag generation, and the exact file format used by the reference CLI. Reported performance numbers are produced by the built in benchmark tool under documented hardware and compiler settings. All security claims are made strictly within the ideal permutation model following standard sponge and duplex bounds, and no stronger guarantees are asserted for the concrete permutation beyond the documented analysis and empirical behavior. The implementation aims for constant time behavior with respect to secret dependent operations, although no formal side channel proof is provided. The project is released under the MIT license, and external cryptanalysis, feedback, and reproducibility checks are explicitly encouraged.

Title: A Single Image and Multimodality Is All You Need for Novel View Synthesis

• Authors: Amirhosein Javadi, Chi-Shiang Gau, Konstantinos D. Polyzos, Tara Javidi
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.17909
• Pdf URL: https://arxiv.org/pdf/2602.17909
• Copy Paste: [[2602.17909]] A Single Image and Multimodality Is All You Need for Novel View Synthesis(https://arxiv.org/abs/2602.17909)
• Keywords: robust, diffusion, generative
• Abstract: Diffusion-based approaches have recently demonstrated strong performance for single-image novel view synthesis by conditioning generative models on geometry inferred from monocular depth estimation. However, in practice, the quality and consistency of the synthesized views are fundamentally limited by the reliability of the underlying depth estimates, which are often fragile under low texture, adverse weather, and occlusion-heavy real-world conditions. In this work, we show that incorporating sparse multimodal range measurements provides a simple yet effective way to overcome these limitations. We introduce a multimodal depth reconstruction framework that leverages extremely sparse range sensing data, such as automotive radar or LiDAR, to produce dense depth maps that serve as robust geometric conditioning for diffusion-based novel view synthesis. Our approach models depth in an angular domain using a localized Gaussian Process formulation, enabling computationally efficient inference while explicitly quantifying uncertainty in regions with limited observations. The reconstructed depth and uncertainty are used as a drop-in replacement for monocular depth estimators in existing diffusion-based rendering pipelines, without modifying the generative model itself. Experiments on real-world multimodal driving scenes demonstrate that replacing vision-only depth with our sparse range-based reconstruction substantially improves both geometric consistency and visual quality in single-image novel-view video generation. These results highlight the importance of reliable geometric priors for diffusion-based view synthesis and demonstrate the practical benefits of multimodal sensing even at extreme levels of sparsity.

Title: Condition-Gated Reasoning for Context-Dependent Biomedical Question Answering

• Authors: Jash Rajesh Parekh, Wonbin Kweon, Joey Chan, Rezarta Islamaj, Robert Leaman, Pengcheng Jiang, Chih-Hsuan Wei, Zhizheng Wang, Zhiyong Lu, Jiawei Han
• Subjects: cs.CL, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17911
• Pdf URL: https://arxiv.org/pdf/2602.17911
• Copy Paste: [[2602.17911]] Condition-Gated Reasoning for Context-Dependent Biomedical Question Answering(https://arxiv.org/abs/2602.17911)
• Keywords: robust
• Abstract: Current biomedical question answering (QA) systems often assume that medical knowledge applies uniformly, yet real-world clinical reasoning is inherently conditional: nearly every decision depends on patient-specific factors such as comorbidities and contraindications. Existing benchmarks do not evaluate such conditional reasoning, and retrieval-augmented or graph-based methods lack explicit mechanisms to ensure that retrieved knowledge is applicable to given context. To address this gap, we propose CondMedQA, the first benchmark for conditional biomedical QA, consisting of multi-hop questions whose answers vary with patient conditions. Furthermore, we propose Condition-Gated Reasoning (CGR), a novel framework that constructs condition-aware knowledge graphs and selectively activates or prunes reasoning paths based on query conditions. Our findings show that CGR more reliably selects condition-appropriate answers while matching or exceeding state-of-the-art performance on biomedical QA benchmarks, highlighting the importance of explicitly modeling conditionality for robust medical reasoning.

Title: ZACH-ViT: Regime-Dependent Inductive Bias in Compact Vision Transformers for Medical Imaging

• Authors: Athanasios Angelakis
• Subjects: cs.CV, cs.LG, eess.IV
• Abstract URL: https://arxiv.org/abs/2602.17929
• Pdf URL: https://arxiv.org/pdf/2602.17929
• Copy Paste: [[2602.17929]] ZACH-ViT: Regime-Dependent Inductive Bias in Compact Vision Transformers for Medical Imaging(https://arxiv.org/abs/2602.17929)
• Keywords: transformer
• Abstract: Vision Transformers rely on positional embeddings and class tokens that encode fixed spatial priors. While effective for natural images, these priors may hinder generalization when spatial layout is weakly informative or inconsistent, a frequent condition in medical imaging and edge-deployed clinical systems. We introduce ZACH-ViT (Zero-token Adaptive Compact Hierarchical Vision Transformer), a compact Vision Transformer that removes both positional embeddings and the [CLS] token, achieving permutation invariance through global average pooling over patch representations. The term "Zero-token" specifically refers to removing the dedicated [CLS] aggregation token and positional embeddings; patch tokens remain unchanged and are processed normally. Adaptive residual projections preserve training stability in compact configurations while maintaining a strict parameter budget. Evaluation is performed across seven MedMNIST datasets spanning binary and multi-class tasks under a strict few-shot protocol (50 samples per class, fixed hyperparameters, five random seeds). The empirical analysis demonstrates regime-dependent behavior: ZACH-ViT (0.25M parameters, trained from scratch) achieves its strongest advantage on BloodMNIST and remains competitive with TransMIL on PathMNIST, while its relative advantage decreases on datasets with strong anatomical priors (OCTMNIST, OrganAMNIST), consistent with the architectural hypothesis. These findings support the view that aligning architectural inductive bias with data structure can be more important than pursuing universal benchmark dominance. Despite its minimal size and lack of pretraining, ZACH-ViT achieves competitive performance while maintaining sub-second inference times, supporting deployment in resource-constrained clinical environments. Code and models are available at this https URL.

Title: MIRA: Memory-Integrated Reinforcement Learning Agent with Limited LLM Guidance

• Authors: Narjes Nourzad, Carlee Joe-Wong
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17930
• Pdf URL: https://arxiv.org/pdf/2602.17930
• Copy Paste: [[2602.17930]] MIRA: Memory-Integrated Reinforcement Learning Agent with Limited LLM Guidance(https://arxiv.org/abs/2602.17930)
• Keywords: large language model
• Abstract: Reinforcement learning (RL) agents often suffer from high sample complexity in sparse or delayed reward settings due to limited prior structure. Large language models (LLMs) can provide subgoal decompositions, plausible trajectories, and abstract priors that facilitate early learning. However, heavy reliance on LLM supervision introduces scalability constraints and dependence on potentially unreliable signals. We propose MIRA (Memory-Integrated Reinforcement Learning Agent), which incorporates a structured, evolving memory graph to guide early training. The graph stores decision-relevant information, including trajectory segments and subgoal structures, and is constructed from both the agent's high-return experiences and LLM outputs. This design amortizes LLM queries into a persistent memory rather than requiring continuous real-time supervision. From this memory graph, we derive a utility signal that softly adjusts advantage estimation to influence policy updates without modifying the underlying reward function. As training progresses, the agent's policy gradually surpasses the initial LLM-derived priors, and the utility term decays, preserving standard convergence guarantees. We provide theoretical analysis showing that utility-based shaping improves early-stage learning in sparse-reward environments. Empirically, MIRA outperforms RL baselines and achieves returns comparable to approaches that rely on frequent LLM supervision, while requiring substantially fewer online LLM queries. Project webpage: this https URL

Title: Memory-Based Advantage Shaping for LLM-Guided Reinforcement Learning

• Authors: Narjes Nourzad, Carlee Joe-Wong
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17931
• Pdf URL: https://arxiv.org/pdf/2602.17931
• Copy Paste: [[2602.17931]] Memory-Based Advantage Shaping for LLM-Guided Reinforcement Learning(https://arxiv.org/abs/2602.17931)
• Keywords: large language model
• Abstract: In environments with sparse or delayed rewards, reinforcement learning (RL) incurs high sample complexity due to the large number of interactions needed for learning. This limitation has motivated the use of large language models (LLMs) for subgoal discovery and trajectory guidance. While LLMs can support exploration, frequent reliance on LLM calls raises concerns about scalability and reliability. We address these challenges by constructing a memory graph that encodes subgoals and trajectories from both LLM guidance and the agent's own successful rollouts. From this graph, we derive a utility function that evaluates how closely the agent's trajectories align with prior successful strategies. This utility shapes the advantage function, providing the critic with additional guidance without altering the reward. Our method relies primarily on offline input and only occasional online queries, avoiding dependence on continuous LLM supervision. Preliminary experiments in benchmark environments show improved sample efficiency and faster early learning compared to baseline RL methods, with final returns comparable to methods that require frequent LLM interaction.

Title: Causal Neighbourhood Learning for Invariant Graph Representations

• Authors: Simi Job, Xiaohui Tao, Taotao Cai, Haoran Xie, Jianming Yong
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17934
• Pdf URL: https://arxiv.org/pdf/2602.17934
• Copy Paste: [[2602.17934]] Causal Neighbourhood Learning for Invariant Graph Representations(https://arxiv.org/abs/2602.17934)
• Keywords: robust
• Abstract: Graph data often contain noisy and spurious correlations that mask the true causal relationships, which are essential for enabling graph models to make predictions based on the underlying causal structure of the data. Dependence on spurious connections makes it challenging for traditional Graph Neural Networks (GNNs) to generalize effectively across different graphs. Furthermore, traditional aggregation methods tend to amplify these spurious patterns, limiting model robustness under distribution shifts. To address these issues, we propose Causal Neighbourhood Learning with Graph Neural Networks (CNL-GNN), a novel framework that performs causal interventions on graph structure. CNL-GNN effectively identifies and preserves causally relevant connections and reduces spurious influences through the generation of counterfactual neighbourhoods and adaptive edge perturbation guided by learnable importance masking and an attention-based mechanism. In addition, by combining structural-level interventions with the disentanglement of causal features from confounding factors, the model learns invariant node representations that are robust and generalize well across different graph structures. Our approach improves causal graph learning beyond traditional feature-based methods, resulting in a robust classification model. Extensive experiments on four publicly available datasets, including multiple domain variants of one dataset, demonstrate that CNL-GNN outperforms state-of-the-art GNN models.

Title: Analyzing LLM Instruction Optimization for Tabular Fact Verification

• Authors: Xiaotang Du, Giwon Hong, Wai-Chung Kwan, Rohit Saxena, Ivan Titov, Pasquale Minervini, Emily Allaway
• Subjects: cs.CL, cs.PL
• Abstract URL: https://arxiv.org/abs/2602.17937
• Pdf URL: https://arxiv.org/pdf/2602.17937
• Copy Paste: [[2602.17937]] Analyzing LLM Instruction Optimization for Tabular Fact Verification(https://arxiv.org/abs/2602.17937)
• Keywords: large language model
• Abstract: Instruction optimization provides a lightweight, model-agnostic approach to enhancing the reasoning performance of large language models (LLMs). This paper presents the first systematic comparison of instruction optimization, based on the DSPy optimization framework, for tabular fact verification. We evaluate four out-of-the-box prompting techniques that cover both text-only prompting and code use: direct prediction, Chain-of-Thought (CoT), ReAct with SQL tools, and CodeAct with Python execution. We study three optimizers from the DSPy framework -- COPRO, MiPROv2, and SIMBA -- across four benchmarks and three model families. We find that instruction optimization consistently improves verification accuracy, with MiPROv2 yielding the most stable gains for CoT, and SIMBA providing the largest benefits for ReAct agents, particularly at larger model scales. Behavioral analyses reveal that SIMBA encourages more direct reasoning paths by applying heuristics, thereby improving numerical comparison abilities in CoT reasoning and helping avoid unnecessary tool calls in ReAct agents. Across different prompting techniques, CoT remains effective for tabular fact checking, especially with smaller models. Although ReAct agents built with larger models can achieve competitive performance, they require careful instruction optimization.

Title: Optimizing Graph Causal Classification Models: Estimating Causal Effects and Addressing Confounders

• Authors: Simi Job, Xiaohui Tao, Taotao Cai, Haoran Xie, Jianming Yong, Xin Wang
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17941
• Pdf URL: https://arxiv.org/pdf/2602.17941
• Copy Paste: [[2602.17941]] Optimizing Graph Causal Classification Models: Estimating Causal Effects and Addressing Confounders(https://arxiv.org/abs/2602.17941)
• Keywords: robust
• Abstract: Graph data is becoming increasingly prevalent due to the growing demand for relational insights in AI across various domains. Organizations regularly use graph data to solve complex problems involving relationships and connections. Causal learning is especially important in this context, since it helps to understand cause-effect relationships rather than mere associations. Since many real-world systems are inherently causal, graphs can efficiently model these systems. However, traditional graph machine learning methods including graph neural networks (GNNs), rely on correlations and are sensitive to spurious patterns and distribution changes. On the other hand, causal models enable robust predictions by isolating true causal factors, thus making them more stable under such shifts. Causal learning also helps in identifying and adjusting for confounders, ensuring that predictions reflect true causal relationships and remain accurate even under interventions. To address these challenges and build models that are robust and causally informed, we propose CCAGNN, a Confounder-Aware causal GNN framework that incorporates causal reasoning into graph learning, supporting counterfactual reasoning and providing reliable predictions in real-world settings. Comprehensive experiments on six publicly available datasets from diverse domains show that CCAGNN consistently outperforms leading state-of-the-art models.

Title: A Geometric Probe of the Accuracy-Robustness Trade-off: Sharp Boundaries in Symmetry-Breaking Dimensional Expansion

• Authors: Yu Bai, Zhe Wang, Jiarui Zhang, Dong-Xiao Zhang, Yinjun Gao, Jun-Jie Zhang
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.17948
• Pdf URL: https://arxiv.org/pdf/2602.17948
• Copy Paste: [[2602.17948]] A Geometric Probe of the Accuracy-Robustness Trade-off: Sharp Boundaries in Symmetry-Breaking Dimensional Expansion(https://arxiv.org/abs/2602.17948)
• Keywords: attack, robust
• Abstract: The trade-off between clean accuracy and adversarial robustness is a pervasive phenomenon in deep learning, yet its geometric origin remains elusive. In this work, we utilize Symmetry-Breaking Dimensional Expansion (SBDE) as a controlled probe to investigate the mechanism underlying this trade-off. SBDE expands input images by inserting constant-valued pixels, which breaks translational symmetry and consistently improves clean accuracy (e.g., from $90.47\%$ to $95.63\%$ on CIFAR-10 with ResNet-18) by reducing parameter degeneracy. However, this accuracy gain comes at the cost of reduced robustness against iterative white-box attacks. By employing a test-time \emph{mask projection} that resets the inserted auxiliary pixels to their training values, we demonstrate that the vulnerability stems almost entirely from the inserted dimensions. The projection effectively neutralizes the attacks and restores robustness, revealing that the model achieves high accuracy by creating \emph{sharp boundaries} (steep loss gradients) specifically along the auxiliary axes. Our findings provide a concrete geometric explanation for the accuracy-robustness paradox: the optimization landscape deepens the basin of attraction to improve accuracy but inevitably erects steep walls along the auxiliary degrees of freedom, creating a fragile sensitivity to off-manifold perturbations.

Title: CUICurate: A GraphRAG-based Framework for Automated Clinical Concept Curation for NLP applications

• Authors: Victoria Blake, Mathew Miller, Jamie Novak, Sze-yuan Ooi, Blanca Gallego
• Subjects: cs.CL, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17949
• Pdf URL: https://arxiv.org/pdf/2602.17949
• Copy Paste: [[2602.17949]] CUICurate: A GraphRAG-based Framework for Automated Clinical Concept Curation for NLP applications(https://arxiv.org/abs/2602.17949)
• Keywords: large language model
• Abstract: Background: Clinical named entity recognition tools commonly map free text to Unified Medical Language System (UMLS) Concept Unique Identifiers (CUIs). For many downstream tasks, however, the clinically meaningful unit is not a single CUI but a concept set comprising related synonyms, subtypes, and supertypes. Constructing such concept sets is labour-intensive, inconsistently performed, and poorly supported by existing tools, particularly for NLP pipelines that operate directly on UMLS CUIs. Methods We present CUICurate, a Graph-based retrieval-augmented generation (GraphRAG) framework for automated UMLS concept set curation. A UMLS knowledge graph (KG) was constructed and embedded for semantic retrieval. For each target concept, candidate CUIs were retrieved from the KG, followed by large language model (LLM) filtering and classification steps comparing two LLMs (GPT-5 and GPT-5-mini). The framework was evaluated on five lexically heterogeneous clinical concepts against a manually curated benchmark and gold-standard concept sets. Results Across all concepts, CUICurate produced substantially larger and more complete concept sets than the manual benchmarks whilst matching human precision. Comparisons between the two LLMs found that GPT-5-mini achieved higher recall during filtering, while GPT-5 produced classifications that more closely aligned with clinician judgements. Outputs were stable across repeated runs and computationally inexpensive. Conclusions CUICurate offers a scalable and reproducible approach to support UMLS concept set curation that substantially reduces manual effort. By integrating graph-based retrieval with LLM reasoning, the framework produces focused candidate concept sets that can be adapted to clinical NLP pipelines for different phenotyping and analytic requirements.

Title: PenTiDef: Enhancing Privacy and Robustness in Decentralized Federated Intrusion Detection Systems against Poisoning Attacks

• Authors: Phan The Duy, Nghi Hoang Khoa, Nguyen Tran Anh Quan, Luong Ha Tien, Ngo Duc Hoang Son, Van-Hau Pham
• Subjects: cs.CR, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17973
• Pdf URL: https://arxiv.org/pdf/2602.17973
• Copy Paste: [[2602.17973]] PenTiDef: Enhancing Privacy and Robustness in Decentralized Federated Intrusion Detection Systems against Poisoning Attacks(https://arxiv.org/abs/2602.17973)
• Keywords: secure, security, privacy, protect, defense, attack, robust, federate
• Abstract: The increasing deployment of Federated Learning (FL) in Intrusion Detection Systems (IDS) introduces new challenges related to data privacy, centralized coordination, and susceptibility to poisoning attacks. While significant research has focused on protecting traditional FL-IDS with centralized aggregation servers, there remains a notable gap in addressing the unique challenges of decentralized FL-IDS (DFL-IDS). This study aims to address the limitations of traditional centralized FL-IDS by proposing a novel defense framework tailored for the decentralized FL-IDS architecture, with a focus on privacy preservation and robustness against poisoning attacks. We propose PenTiDef, a privacy-preserving and robust defense framework for DFL-IDS, which incorporates Distributed Differential Privacy (DDP) to protect data confidentiality and utilizes latent space representations (LSR) derived from neural networks to detect malicious updates in the decentralized model aggregation context. To eliminate single points of failure and enhance trust without a centralized aggregation server, PenTiDef employs a blockchain-based decentralized coordination mechanism that manages model aggregation, tracks update history, and supports trust enforcement through smart contracts. Experimental results on CIC-IDS2018 and Edge-IIoTSet demonstrate that PenTiDef consistently outperforms existing defenses (e.g., FLARE, FedCC) across various attack scenarios and data distributions. These findings highlight the potential of PenTiDef as a scalable and secure framework for deploying DFL-based IDS in adversarial environments. By leveraging privacy protection, malicious behavior detection in hidden data, and working without a central server, it provides a useful security solution against real-world attacks from untrust participants.

Title: Generating adversarial inputs for a graph neural network model of AC power flow

• Authors: Robert Parker
• Subjects: cs.LG, eess.SY
• Abstract URL: https://arxiv.org/abs/2602.17975
• Pdf URL: https://arxiv.org/pdf/2602.17975
• Copy Paste: [[2602.17975]] Generating adversarial inputs for a graph neural network model of AC power flow(https://arxiv.org/abs/2602.17975)
• Keywords: robust
• Abstract: This work formulates and solves optimization problems to generate input points that yield high errors between a neural network's predicted AC power flow solution and solutions to the AC power flow equations. We demonstrate this capability on an instance of the CANOS-PF graph neural network model, as implemented by the PF$\Delta$ benchmark library, operating on a 14-bus test grid. Generated adversarial points yield errors as large as 3.4 per-unit in reactive power and 0.08 per-unit in voltage magnitude. When minimizing the perturbation from a training point necessary to satisfy adversarial constraints, we find that the constraints can be met with as little as an 0.04 per-unit perturbation in voltage magnitude on a single bus. This work motivates the development of rigorous verification and robust training methods for neural network surrogate models of AC power flow.

Title: Decomposing Retrieval Failures in RAG for Long-Document Financial Question Answering

• Authors: Amine Kobeissi, Philippe Langlais
• Subjects: cs.CL, cs.IR
• Abstract URL: https://arxiv.org/abs/2602.17981
• Pdf URL: https://arxiv.org/pdf/2602.17981
• Copy Paste: [[2602.17981]] Decomposing Retrieval Failures in RAG for Long-Document Financial Question Answering(https://arxiv.org/abs/2602.17981)
• Keywords: generative
• Abstract: Retrieval-augmented generation is increasingly used for financial question answering over long regulatory filings, yet reliability depends on retrieving the exact context needed to justify answers in high stakes settings. We study a frequent failure mode in which the correct document is retrieved but the page or chunk that contains the answer is missed, leading the generator to extrapolate from incomplete context. Despite its practical significance, this within-document retrieval failure mode has received limited systematic attention in the Financial Question Answering (QA) literature. We evaluate retrieval at multiple levels of granularity, document, page, and chunk level, and introduce an oracle based analysis to provide empirical upper bounds on retrieval and generative performance. On a 150 question subset of FinanceBench, we reproduce and compare diverse retrieval strategies including dense, sparse, hybrid, and hierarchical methods with reranking and query reformulation. Across methods, gains in document discovery tend to translate into stronger page recall, yet oracle performance still suggests headroom for page and chunk level retrieval. To target this gap, we introduce a domain fine-tuned page scorer that treats pages as an intermediate retrieval unit between documents and chunks. Unlike prior passage-based hierarchical retrieval, we fine-tune a bi-encoder specifically for page-level relevance on financial filings, exploiting the semantic coherence of pages. Overall, our results demonstrate a significant improvement in page recall and chunk retrieval.

Title: Turbo Connection: Reasoning as Information Flow from Higher to Lower Layers

• Authors: Mohan Tang, Sidi Lu
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.17993
• Pdf URL: https://arxiv.org/pdf/2602.17993
• Copy Paste: [[2602.17993]] Turbo Connection: Reasoning as Information Flow from Higher to Lower Layers(https://arxiv.org/abs/2602.17993)
• Keywords: transformer
• Abstract: Complex problems, whether in math, logic, or planning, are solved by humans through a sequence of steps where the result of one step informs the next. In this work, we adopt the perspective that the reasoning power of Transformers is fundamentally limited by a fixed maximum number of steps along any latent path of computation. To address this, we introduce Turbo Connection (TurboConn), a novel architecture that overcomes the fixed-depth constraint by routing multiple residual connections from the higher-layer hidden states of each token $t$ to the lower layers of token $t+1$. Fine-tuning pre-trained LLMs with our method not only yields accuracy gains of 0.9% to over 10% on benchmarks like GSM8K, Parity, and multi-step arithmetic, but also demonstrates that the density of these backward connections is critical; our dense interaction significantly outperforms "sparse" alternatives that only pass a single hidden state or vector. Notably, TurboConn can be integrated into pre-trained LLMs to overcome task-specific plateaus: while a fine-tuned Qwen-3-1.7B achieves only 53.78% on Parity, adding our architectural modification enables the model to reach 100% accuracy, all without the necessity to retrain the full model from scratch or sophisticated curriculum learning. Our results provide strong empirical evidence that the depth of the computational path is a key factor in reasoning ability, also offering a new mechanism to enhance LLMs without significantly affecting generation latency.

Title: Asynchronous Heavy-Tailed Optimization

• Authors: Junfei Sun, Dixi Yao, Xuchen Gong, Tahseen Rabbani, Manzil Zaheer, Tian Li
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18002
• Pdf URL: https://arxiv.org/pdf/2602.18002
• Copy Paste: [[2602.18002]] Asynchronous Heavy-Tailed Optimization(https://arxiv.org/abs/2602.18002)
• Keywords: robust, transformer
• Abstract: Heavy-tailed stochastic gradient noise, commonly observed in transformer models, can destabilize the optimization process. Recent works mainly focus on developing and understanding approaches to address heavy-tailed noise in the centralized or distributed, synchronous setting, leaving the interactions between such noise and asynchronous optimization underexplored. In this work, we investigate two communication schemes that handle stragglers with asynchronous updates in the presence of heavy-tailed gradient noise. We propose and theoretically analyze algorithmic modifications based on delay-aware learning rate scheduling and delay compensation to enhance the performance of asynchronous algorithms. Our convergence guarantees under heavy-tailed noise match the rate of the synchronous counterparts and improve delay tolerance compared with existing asynchronous approaches. Empirically, our approaches outperform prior synchronous and asynchronous methods in terms of accuracy/runtime trade-offs and are more robust to hyperparameters in both image and language tasks.

Title: MUOT_3M: A 3 Million Frame Multimodal Underwater Benchmark and the MUTrack Tracking Method

• Authors: Ahsan Baidar Bakht, Mohamad Alansari, Muhayy Ud Din, Muzammal Naseer, Sajid Javed, Irfan Hussain, Jiri Matas, Arif Mahmood
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18006
• Pdf URL: https://arxiv.org/pdf/2602.18006
• Copy Paste: [[2602.18006]] MUOT_3M: A 3 Million Frame Multimodal Underwater Benchmark and the MUTrack Tracking Method(https://arxiv.org/abs/2602.18006)
• Keywords: robust
• Abstract: Underwater Object Tracking (UOT) is crucial for efficient marine robotics, large scale ecological monitoring, and ocean exploration; however, progress has been hindered by the scarcity of large, multimodal, and diverse datasets. Existing benchmarks remain small and RGB only, limiting robustness under severe color distortion, turbidity, and low visibility conditions. We introduce MUOT_3M, the first pseudo multimodal UOT benchmark comprising 3 million frames from 3,030 videos (27.8h) annotated with 32 tracking attributes, 677 fine grained classes, and synchronized RGB, estimated enhanced RGB, estimated depth, and language modalities validated by a marine biologist. Building upon MUOT_3M, we propose MUTrack, a SAM-based multimodal to unimodal tracker featuring visual geometric alignment, vision language fusion, and four level knowledge distillation that transfers multimodal knowledge into a unimodal student model. Extensive evaluations across five UOT benchmarks demonstrate that MUTrack achieves up to 8.40% higher AUC and 7.80% higher precision than the strongest SOTA baselines while running at 24 FPS. MUOT_3M and MUTrack establish a new foundation for scalable, multimodally trained yet practically deployable underwater tracking.

Title: DeepSVU: Towards In-depth Security-oriented Video Understanding via Unified Physical-world Regularized MoE

• Authors: Yujie Jin, Wenxin Zhang, Jingjing Wang, Guodong Zhou
• Subjects: cs.CV, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18019
• Pdf URL: https://arxiv.org/pdf/2602.18019
• Copy Paste: [[2602.18019]] DeepSVU: Towards In-depth Security-oriented Video Understanding via Unified Physical-world Regularized MoE(https://arxiv.org/abs/2602.18019)
• Keywords: security
• Abstract: In the literature, prior research on Security-oriented Video Understanding (SVU) has predominantly focused on detecting and localize the threats (e.g., shootings, robberies) in videos, while largely lacking the effective capability to generate and evaluate the threat causes. Motivated by these gaps, this paper introduces a new chat paradigm SVU task, i.e., In-depth Security-oriented Video Understanding (DeepSVU), which aims to not only identify and locate the threats but also attribute and evaluate the causes threatening segments. Furthermore, this paper reveals two key challenges in the proposed task: 1) how to effectively model the coarse-to-fine physical-world information (e.g., human behavior, object interactions and background context) to boost the DeepSVU task; and 2) how to adaptively trade off these factors. To tackle these challenges, this paper proposes a new Unified Physical-world Regularized MoE (UPRM) approach. Specifically, UPRM incorporates two key components: the Unified Physical-world Enhanced MoE (UPE) Block and the Physical-world Trade-off Regularizer (PTR), to address the above two challenges, respectively. Extensive experiments conduct on our DeepSVU instructions datasets (i.e., UCF-C instructions and CUVA instructions) demonstrate that UPRM outperforms several advanced Video-LLMs as well as non-VLM approaches. Such this http URL justify the importance of the coarse-to-fine physical-world information in the DeepSVU task and demonstrate the effectiveness of our UPRM in capturing such information.

Title: Dual-Channel Attention Guidance for Training-Free Image Editing Control in Diffusion Transformers

• Authors: Guandong Li, Mengxia Ye
• Subjects: cs.CV, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18022
• Pdf URL: https://arxiv.org/pdf/2602.18022
• Copy Paste: [[2602.18022]] Dual-Channel Attention Guidance for Training-Free Image Editing Control in Diffusion Transformers(https://arxiv.org/abs/2602.18022)
• Keywords: diffusion, transformer
• Abstract: Training-free control over editing intensity is a critical requirement for diffusion-based image editing models built on the Diffusion Transformer (DiT) architecture. Existing attention manipulation methods focus exclusively on the Key space to modulate attention routing, leaving the Value space -- which governs feature aggregation -- entirely unexploited. In this paper, we first reveal that both Key and Value projections in DiT's multi-modal attention layers exhibit a pronounced bias-delta structure, where token embeddings cluster tightly around a layer-specific bias vector. Building on this observation, we propose Dual-Channel Attention Guidance (DCAG), a training-free framework that simultaneously manipulates both the Key channel (controlling where to attend) and the Value channel (controlling what to aggregate). We provide a theoretical analysis showing that the Key channel operates through the nonlinear softmax function, acting as a coarse control knob, while the Value channel operates through linear weighted summation, serving as a fine-grained complement. Together, the two-dimensional parameter space $(\delta_k, \delta_v)$ enables more precise editing-fidelity trade-offs than any single-channel method. Extensive experiments on the PIE-Bench benchmark (700 images, 10 editing categories) demonstrate that DCAG consistently outperforms Key-only guidance across all fidelity metrics, with the most significant improvements observed in localized editing tasks such as object deletion (4.9% LPIPS reduction) and object addition (3.2% LPIPS reduction).

Title: Spatio-temporal Decoupled Knowledge Compensator for Few-Shot Action Recognition

• Authors: Hongyu Qu, Xiangbo Shu, Rui Yan, Hailiang Gao, Wenguan Wang, Jinhui Tang
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18043
• Pdf URL: https://arxiv.org/pdf/2602.18043
• Copy Paste: [[2602.18043]] Spatio-temporal Decoupled Knowledge Compensator for Few-Shot Action Recognition(https://arxiv.org/abs/2602.18043)
• Keywords: large language model
• Abstract: Few-Shot Action Recognition (FSAR) is a challenging task that requires recognizing novel action categories with a few labeled videos. Recent works typically apply semantically coarse category names as auxiliary contexts to guide the learning of discriminative visual features. However, such context provided by the action names is too limited to provide sufficient background knowledge for capturing novel spatial and temporal concepts in actions. In this paper, we propose DiST, an innovative Decomposition-incorporation framework for FSAR that makes use of decoupled Spatial and Temporal knowledge provided by large language models to learn expressive multi-granularity prototypes. In the decomposition stage, we decouple vanilla action names into diverse spatio-temporal attribute descriptions (action-related knowledge). Such commonsense knowledge complements semantic contexts from spatial and temporal perspectives. In the incorporation stage, we propose Spatial/Temporal Knowledge Compensators (SKC/TKC) to discover discriminative object-level and frame-level prototypes, respectively. In SKC, object-level prototypes adaptively aggregate important patch tokens under the guidance of spatial knowledge. Moreover, in TKC, frame-level prototypes utilize temporal attributes to assist in inter-frame temporal relation modeling. These learned prototypes thus provide transparency in capturing fine-grained spatial details and diverse temporal patterns. Experimental results show DiST achieves state-of-the-art results on five standard FSAR datasets.

Title: CityGuard: Graph-Aware Private Descriptors for Bias-Resilient Identity Search Across Urban Cameras

• Authors: Rong Fu, Wenxin Zhang, Yibo Meng, Jia Yee Tan, Jiaxuan Lu, Rui Lu, Jiekai Wu, Zhaolu Kang, Simon Fong
• Subjects: cs.CV, cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18047
• Pdf URL: https://arxiv.org/pdf/2602.18047
• Copy Paste: [[2602.18047]] CityGuard: Graph-Aware Private Descriptors for Bias-Resilient Identity Search Across Urban Cameras(https://arxiv.org/abs/2602.18047)
• Keywords: secure, privacy, protect, robust, transformer
• Abstract: City-scale person re-identification across distributed cameras must handle severe appearance changes from viewpoint, occlusion, and domain shift while complying with data protection rules that prevent sharing raw imagery. We introduce CityGuard, a topology-aware transformer for privacy-preserving identity retrieval in decentralized surveillance. The framework integrates three components. A dispersion-adaptive metric learner adjusts instance-level margins according to feature spread, increasing intra-class compactness. Spatially conditioned attention injects coarse geometry, such as GPS or deployment floor plans, into graph-based self-attention to enable projectively consistent cross-view alignment using only coarse geometric priors without requiring survey-grade calibration. Differentially private embedding maps are coupled with compact approximate indexes to support secure and cost-efficient deployment. Together these designs produce descriptors robust to viewpoint variation, occlusion, and domain shifts, and they enable a tunable balance between privacy and utility under rigorous differential-privacy accounting. Experiments on Market-1501 and additional public benchmarks, complemented by database-scale retrieval studies, show consistent gains in retrieval precision and query throughput over strong baselines, confirming the practicality of the framework for privacy-critical urban identity matching.

Title: Continual-NExT: A Unified Comprehension And Generation Continual Learning Framework

• Authors: Jingyang Qiao, Zhizhong Zhang, Xin Tan, Jingyu Gong, Yanyun Qu, Yuan Xie
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18055
• Pdf URL: https://arxiv.org/pdf/2602.18055
• Copy Paste: [[2602.18055]] Continual-NExT: A Unified Comprehension And Generation Continual Learning Framework(https://arxiv.org/abs/2602.18055)
• Keywords: large language model
• Abstract: Dual-to-Dual MLLMs refer to Multimodal Large Language Models, which can enable unified multimodal comprehension and generation through text and image modalities. Although exhibiting strong instantaneous learning and generalization capabilities, Dual-to-Dual MLLMs still remain deficient in lifelong evolution, significantly affecting continual adaptation to dynamic real-world scenarios. One of the challenges is that learning new tasks inevitably destroys the learned knowledge. Beyond traditional catastrophic forgetting, Dual-to-Dual MLLMs face other challenges, including hallucination, instruction unfollowing, and failures in cross-modal knowledge transfer. However, no standardized continual learning framework for Dual-to-Dual MLLMs has been established yet, leaving these challenges unexplored. Thus, in this paper, we establish Continual-NExT, a continual learning framework for Dual-to-Dual MLLMs with deliberately-architected evaluation metrics. To improve the continual learning capability of Dual-to-Dual MLLMs, we propose an efficient MAGE (Mixture and Aggregation of General LoRA and Expert LoRA) method to further facilitate knowledge transfer across modalities and mitigate forgetting. Extensive experiments demonstrate that MAGE outperforms other continual learning methods and achieves state-of-the-art performance.

Title: Temporal Consistency-Aware Text-to-Motion Generation

• Authors: Hongsong Wang, Wenjing Yan, Qiuxia Lai, Xin Geng
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18057
• Pdf URL: https://arxiv.org/pdf/2602.18057
• Copy Paste: [[2602.18057]] Temporal Consistency-Aware Text-to-Motion Generation(https://arxiv.org/abs/2602.18057)
• Keywords: robust, transformer
• Abstract: Text-to-Motion (T2M) generation aims to synthesize realistic human motion sequences from natural language descriptions. While two-stage frameworks leveraging discrete motion representations have advanced T2M research, they often neglect cross-sequence temporal consistency, i.e., the shared temporal structures present across different instances of the same action. This leads to semantic misalignments and physically implausible motions. To address this limitation, we propose TCA-T2M, a framework for temporal consistency-aware T2M generation. Our approach introduces a temporal consistency-aware spatial VQ-VAE (TCaS-VQ-VAE) for cross-sequence temporal alignment, coupled with a masked motion transformer for text-conditioned motion generation. Additionally, a kinematic constraint block mitigates discretization artifacts to ensure physical plausibility. Experiments on HumanML3D and KIT-ML benchmarks demonstrate that TCA-T2M achieves state-of-the-art performance, highlighting the importance of temporal consistency in robust and coherent T2M generation.

Title: Deepmechanics

• Authors: Abhay Shinde, Aryan Amit Barsainyan, Jose Siguenza, Ankita Vaishnobi Bisoi, Rakshit Kr. Singh, Bharath Ramsundar
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18060
• Pdf URL: https://arxiv.org/pdf/2602.18060
• Copy Paste: [[2602.18060]] Deepmechanics(https://arxiv.org/abs/2602.18060)
• Keywords: robust
• Abstract: Physics-informed deep learning models have emerged as powerful tools for learning dynamical systems. These models directly encode physical principles into network architectures. However, systematic benchmarking of these approaches across diverse physical phenomena remains limited, particularly in conservative and dissipative systems. In addition, benchmarking that has been done thus far does not integrate out full trajectories to check stability. In this work, we benchmark three prominent physics-informed architectures such as Hamiltonian Neural Networks (HNN), Lagrangian Neural Networks (LNN), and Symplectic Recurrent Neural Networks (SRNN) using the DeepChem framework, an open-source scientific machine learning library. We evaluate these models on six dynamical systems spanning classical conservative mechanics (mass-spring system, simple pendulum, double pendulum, and three-body problem, spring-pendulum) and non-conservative systems with contact (bouncing ball). We evaluate models by computing error on predicted trajectories and evaluate error both quantitatively and qualitatively. We find that all benchmarked models struggle to maintain stability for chaotic or nonconservative systems. Our results suggest that more research is needed for physics-informed deep learning models to learn robust models of classical mechanical systems.

Title: Distributed Security: From Isolated Properties to Synergistic Trust

• Authors: Minghui Xu
• Subjects: cs.CR, cs.DC
• Abstract URL: https://arxiv.org/abs/2602.18063
• Pdf URL: https://arxiv.org/pdf/2602.18063
• Copy Paste: [[2602.18063]] Distributed Security: From Isolated Properties to Synergistic Trust(https://arxiv.org/abs/2602.18063)
• Keywords: security, privacy
• Abstract: Over the past four decades, distributed security has undergone a remarkable transformation -- from crash-fault tolerant protocols designed for controlled environments to sophisticated Byzantine-resilient architectures operating in open, adversarial settings. This vision paper examines this evolution and argues for a fundamental shift in how we approach distributed security: from studying individual security properties in isolation to understanding their synergistic combinations. We begin by conclude four foundational properties, \textit{agreement, consistency, privacy, verifiability, accountability}. We trace their theoretical origins and practical maturation. We then demonstrate how the frontier of research now lies at the intersection of these properties, where their fusion creates capabilities that neither property could achieve alone. Looking forward, we identify critical research challenges: discovering new security properties driven by emerging applications, developing systematic frameworks for property convergence, managing the computational overhead of cryptographic primitives in high-performance consensus layers, and addressing post-quantum and human-factor challenges. The future of distributed security lies not in improving individual properties, but in understanding and harnessing their synergies to build a singular fabric of trust.

Title: 3DMedAgent: Unified Perception-to-Understanding for 3D Medical Analysis

• Authors: Ziyue Wang, Linghan Cai, Chang Han Low, Haofeng Liu, Junde Wu, Jingyu Wang, Rui Wang, Lei Song, Jiang Bian, Jingjing Fu, Yueming Jin
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18064
• Pdf URL: https://arxiv.org/pdf/2602.18064
• Copy Paste: [[2602.18064]] 3DMedAgent: Unified Perception-to-Understanding for 3D Medical Analysis(https://arxiv.org/abs/2602.18064)
• Keywords: large language model
• Abstract: 3D CT analysis spans a continuum from low-level perception to high-level clinical understanding. Existing 3D-oriented analysis methods adopt either isolated task-specific modeling or task-agnostic end-to-end paradigms to produce one-hop outputs, impeding the systematic accumulation of perceptual evidence for downstream reasoning. In parallel, recent multimodal large language models (MLLMs) exhibit improved visual perception and can integrate visual and textual information effectively, yet their predominantly 2D-oriented designs fundamentally limit their ability to perceive and analyze volumetric medical data. To bridge this gap, we propose 3DMedAgent, a unified agent that enables 2D MLLMs to perform general 3D CT analysis without 3D-specific fine-tuning. 3DMedAgent coordinates heterogeneous visual and textual tools through a flexible MLLM agent, progressively decomposing complex 3D analysis into tractable subtasks that transition from global to regional views, from 3D volumes to informative 2D slices, and from visual evidence to structured textual representations. Central to this design, 3DMedAgent maintains a long-term structured memory that aggregates intermediate tool outputs and supports query-adaptive, evidence-driven multi-step reasoning. We further introduce the DeepChestVQA benchmark for evaluating unified perception-to-understanding capabilities in 3D thoracic imaging. Experiments across over 40 tasks demonstrate that 3DMedAgent consistently outperforms general, medical, and 3D-specific MLLMs, highlighting a scalable path toward general-purpose 3D clinical this http URL and data are available at \href{this https URL}{this https URL}.

Title: Faster Training, Fewer Labels: Self-Supervised Pretraining for Fine-Grained BEV Segmentation

• Authors: Daniel Busch, Christian Bohn, Thomas Kurbiel, Klaus Friedrichs, Richard Meyes, Tobias Meisen
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18066
• Pdf URL: https://arxiv.org/pdf/2602.18066
• Copy Paste: [[2602.18066]] Faster Training, Fewer Labels: Self-Supervised Pretraining for Fine-Grained BEV Segmentation(https://arxiv.org/abs/2602.18066)
• Keywords: segmentation
• Abstract: Dense Bird's Eye View (BEV) semantic maps are central to autonomous driving, yet current multi-camera methods depend on costly, inconsistently annotated BEV ground truth. We address this limitation with a two-phase training strategy for fine-grained road marking segmentation that removes full supervision during pretraining and halves the amount of training data during fine-tuning while still outperforming the comparable supervised baseline model. During the self-supervised pretraining, BEVFormer predictions are differentiably reprojected into the image plane and trained against multi-view semantic pseudo-labels generated by the widely used semantic segmentation model Mask2Former. A temporal loss encourages consistency across frames. The subsequent supervised fine-tuning phase requires only 50% of the dataset and significantly less training time. With our method, the fine-tuning benefits from rich priors learned during pretraining boosting the performance and BEV segmentation quality (up to +2.5pp mIoU over the fully supervised baseline) on nuScenes. It simultaneously halves the usage of annotation data and reduces total training time by up to two thirds. The results demonstrate that differentiable reprojection plus camera perspective pseudo labels yields transferable BEV features and a scalable path toward reduced-label autonomous perception.

Title: Dynamic Deception: When Pedestrians Team Up to Fool Autonomous Cars

• Authors: Masoud Jamshidiyan Tehrani, Marco Gabriel, Jinhan Kim, Paolo Tonella
• Subjects: cs.CR, cs.RO
• Abstract URL: https://arxiv.org/abs/2602.18079
• Pdf URL: https://arxiv.org/pdf/2602.18079
• Copy Paste: [[2602.18079]] Dynamic Deception: When Pedestrians Team Up to Fool Autonomous Cars(https://arxiv.org/abs/2602.18079)
• Keywords: attack, robust
• Abstract: Many adversarial attacks on autonomous-driving perception models fail to cause system-level failures once deployed in a full driving stack. The main reason for such ineffectiveness is that once deployed in a system (e.g., within a simulator), attacks tend to be spatially or temporally short-lived, due to the vehicle's dynamics, hence rarely influencing the vehicle behaviour. In this paper, we address both limitations by introducing a system-level attack in which multiple dynamic elements (e.g., two pedestrians) carry adversarial patches (e.g., on cloths) and jointly amplify their effect through coordination and motion. We evaluate our attacks in the CARLA simulator using a state-of-the-art autonomous driving agent. At the system level, single-pedestrian attacks fail in all runs (out of 10), while dynamic collusion by two pedestrians induces full vehicle stops in up to 50\% of runs, with static collusion yielding no successful attack at all. These results show that system-level failures arise only when adversarial signals persist over time and are amplified through coordinated actors, exposing a gap between model-level robustness and end-to-end safety.

Title: AndroWasm: an Empirical Study on Android Malware Obfuscation through WebAssembly

• Authors: Diego Soi, Silvia Lucia Sanna, Lorenzo Pisu, Leonardo Regano, Giorgio Giacinto
• Subjects: cs.CR
• Abstract URL: https://arxiv.org/abs/2602.18082
• Pdf URL: https://arxiv.org/pdf/2602.18082
• Copy Paste: [[2602.18082]] AndroWasm: an Empirical Study on Android Malware Obfuscation through WebAssembly(https://arxiv.org/abs/2602.18082)
• Keywords: attack, steal
• Abstract: In recent years, stealthy Android malware has increasingly adopted sophisticated techniques to bypass automatic detection mechanisms and harden manual analysis. Adversaries typically rely on obfuscation, anti-repacking, steganography, poisoning, and evasion techniques to AI-based tools, and in-memory execution to conceal malicious functionality. In this paper, we investigate WebAssembly (Wasm) as a novel technique for hiding malicious payloads and evading traditional static analysis and signature-matching mechanisms. While Wasm is typically employed to render specific gaming activities and interact with the native components in web browsers, we provide an in-depth analysis on the mechanisms Android may employ to include Wasm modules in its execution pipeline. Additionally, we provide Proofs-of-Concept to demonstrate a threat model in which an attacker embeds and executes malicious routines, effectively bypassing IoC detection by industrial state-of-the-art tools, like VirusTotal and MobSF.

Title: Balancing Symmetry and Efficiency in Graph Flow Matching

• Authors: Benjamin Honoré, Alba Carballo-Castro, Yiming Qin, Pascal Frossard
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18084
• Pdf URL: https://arxiv.org/pdf/2602.18084
• Copy Paste: [[2602.18084]] Balancing Symmetry and Efficiency in Graph Flow Matching(https://arxiv.org/abs/2602.18084)
• Keywords: generative
• Abstract: Equivariance is central to graph generative models, as it ensures the model respects the permutation symmetry of graphs. However, strict equivariance can increase computational cost due to added architectural constraints, and can slow down convergence because the model must be consistent across a large space of possible node permutations. We study this trade-off for graph generative models. Specifically, we start from an equivariant discrete flow-matching model, and relax its equivariance during training via a controllable symmetry modulation scheme based on sinusoidal positional encodings and node permutations. Experiments first show that symmetry-breaking can accelerate early training by providing an easier learning signal, but at the expense of encouraging shortcut solutions that can cause overfitting, where the model repeatedly generates graphs that are duplicates of the training set. On the contrary, properly modulating the symmetry signal can delay overfitting while accelerating convergence, allowing the model to reach stronger performance with $19\%$ of the baseline training epochs.

Title: DohaScript: A Large-Scale Multi-Writer Dataset for Continuous Handwritten Hindi Text

• Authors: Kunwar Arpit Singh, Ankush Prakash, Haroon R Lone
• Subjects: cs.CV, cs.AI, cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18089
• Pdf URL: https://arxiv.org/pdf/2602.18089
• Copy Paste: [[2602.18089]] DohaScript: A Large-Scale Multi-Writer Dataset for Continuous Handwritten Hindi Text(https://arxiv.org/abs/2602.18089)
• Keywords: generative
• Abstract: Despite having hundreds of millions of speakers, handwritten Devanagari text remains severely underrepresented in publicly available benchmark datasets. Existing resources are limited in scale, focus primarily on isolated characters or short words, and lack controlled lexical content and writer level diversity, which restricts their utility for modern data driven handwriting analysis. As a result, they fail to capture the continuous, fused, and structurally complex nature of Devanagari handwriting, where characters are connected through a shared shirorekha (horizontal headline) and exhibit rich ligature formations. We introduce DohaScript, a large scale, multi writer dataset of handwritten Hindi text collected from 531 unique contributors. The dataset is designed as a parallel stylistic corpus, in which all writers transcribe the same fixed set of six traditional Hindi dohas (couplets). This controlled design enables systematic analysis of writer specific variation independent of linguistic content, and supports tasks such as handwriting recognition, writer identification, style analysis, and generative modeling. The dataset is accompanied by non identifiable demographic metadata, rigorous quality curation based on objective sharpness and resolution criteria, and page level layout difficulty annotations that facilitate stratified benchmarking. Baseline experiments demonstrate clear quality separation and strong generalization to unseen writers, highlighting the dataset's reliability and practical value. DohaScript is intended to serve as a standardized and reproducible benchmark for advancing research on continuous handwritten Devanagari text in low resource script settings.

Title: Perceived Political Bias in LLMs Reduces Persuasive Abilities

• Authors: Matthew DiGiuseppe, Joshua Robison
• Subjects: cs.CL, cs.AI, cs.CY
• Abstract URL: https://arxiv.org/abs/2602.18092
• Pdf URL: https://arxiv.org/pdf/2602.18092
• Copy Paste: [[2602.18092]] Perceived Political Bias in LLMs Reduces Persuasive Abilities(https://arxiv.org/abs/2602.18092)
• Keywords: attack
• Abstract: Conversational AI has been proposed as a scalable way to correct public misconceptions and spread misinformation. Yet its effectiveness may depend on perceptions of its political neutrality. As LLMs enter partisan conflict, elites increasingly portray them as ideologically aligned. We test whether these credibility attacks reduce LLM-based persuasion. In a preregistered U.S. survey experiment (N=2144), participants completed a three-round conversation with ChatGPT about a personally held economic policy misconception. Compared to a neutral control, a short message indicating that the LLM was biased against the respondent's party attenuated persuasion by 28%. Transcript analysis indicates that the warnings alter the interaction: respondents push back more and engage less receptively. These findings suggest that the persuasive impact of conversational AI is politically contingent, constrained by perceptions of partisan alignment.

Title: Predict to Skip: Linear Multistep Feature Forecasting for Efficient Diffusion Transformers

• Authors: Hanshuai Cui, Zhiqing Tang, Qianli Ma, Zhi Yao, Weijia Jia
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18093
• Pdf URL: https://arxiv.org/pdf/2602.18093
• Copy Paste: [[2602.18093]] Predict to Skip: Linear Multistep Feature Forecasting for Efficient Diffusion Transformers(https://arxiv.org/abs/2602.18093)
• Keywords: diffusion, transformer
• Abstract: Diffusion Transformers (DiT) have emerged as a widely adopted backbone for high-fidelity image and video generation, yet their iterative denoising process incurs high computational costs. Existing training-free acceleration methods rely on feature caching and reuse under the assumption of temporal stability. However, reusing features for multiple steps may lead to latent drift and visual degradation. We observe that model outputs evolve smoothly along much of the diffusion trajectory, enabling principled predictions rather than naive reuse. Based on this insight, we propose \textbf{PrediT}, a training-free acceleration framework that formulates feature prediction as a linear multistep problem. We employ classical linear multistep methods to forecast future model outputs from historical information, combined with a corrector that activates in high-dynamics regions to prevent error accumulation. A dynamic step modulation mechanism adaptively adjusts the prediction horizon by monitoring the feature change rate. Together, these components enable substantial acceleration while preserving generation fidelity. Extensive experiments validate that our method achieves up to $5.54\times$ latency reduction across various DiT-based image and video generation models, while incurring negligible quality degradation.

Title: TempoNet: Slack-Quantized Transformer-Guided Reinforcement Scheduler for Adaptive Deadline-Centric Real-Time Dispatchs

• Authors: Rong Fu, Yibo Meng, Guangzhen Yao, Jiaxuan Lu, Zeyu Zhang, Zhaolu Kang, Ziming Guo, Jia Yee Tan, Xiaojing Du, Simon James Fong
• Subjects: cs.LG, cs.OS, eess.SY
• Abstract URL: https://arxiv.org/abs/2602.18109
• Pdf URL: https://arxiv.org/pdf/2602.18109
• Copy Paste: [[2602.18109]] TempoNet: Slack-Quantized Transformer-Guided Reinforcement Scheduler for Adaptive Deadline-Centric Real-Time Dispatchs(https://arxiv.org/abs/2602.18109)
• Keywords: robust, transformer
• Abstract: Real-time schedulers must reason about tight deadlines under strict compute budgets. We present TempoNet, a reinforcement learning scheduler that pairs a permutation-invariant Transformer with a deep Q-approximation. An Urgency Tokenizer discretizes temporal slack into learnable embeddings, stabilizing value learning and capturing deadline proximity. A latency-aware sparse attention stack with blockwise top-k selection and locality-sensitive chunking enables global reasoning over unordered task sets with near-linear scaling and sub-millisecond inference. A multicore mapping layer converts contextualized Q-scores into processor assignments through masked-greedy selection or differentiable matching. Extensive evaluations on industrial mixed-criticality traces and large multiprocessor settings show consistent gains in deadline fulfillment over analytic schedulers and neural baselines, together with improved optimization stability. Diagnostics include sensitivity analyses for slack quantization, attention-driven policy interpretation, hardware-in-the-loop and kernel micro-benchmarks, and robustness under stress with simple runtime mitigations; we also report sample-efficiency benefits from behavioral-cloning pretraining and compatibility with an actor-critic variant without altering the inference pipeline. These results establish a practical framework for Transformer-based decision making in high-throughput real-time scheduling.

Title: Flow Matching with Injected Noise for Offline-to-Online Reinforcement Learning

• Authors: Yongjae Shin, Jongseong Chae, Jongeui Park, Youngchul Sung
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18117
• Pdf URL: https://arxiv.org/pdf/2602.18117
• Copy Paste: [[2602.18117]] Flow Matching with Injected Noise for Offline-to-Online Reinforcement Learning(https://arxiv.org/abs/2602.18117)
• Keywords: generative
• Abstract: Generative models have recently demonstrated remarkable success across diverse domains, motivating their adoption as expressive policies in reinforcement learning (RL). While they have shown strong performance in offline RL, particularly where the target distribution is well defined, their extension to online fine-tuning has largely been treated as a direct continuation of offline pre-training, leaving key challenges unaddressed. In this paper, we propose Flow Matching with Injected Noise for Offline-to-Online RL (FINO), a novel method that leverages flow matching-based policies to enhance sample efficiency for offline-to-online RL. FINO facilitates effective exploration by injecting noise into policy training, thereby encouraging a broader range of actions beyond those observed in the offline dataset. In addition to exploration-enhanced flow policy training, we combine an entropy-guided sampling mechanism to balance exploration and exploitation, allowing the policy to adapt its behavior throughout online fine-tuning. Experiments across diverse, challenging tasks demonstrate that FINO consistently achieves superior performance under limited online budgets.

Title: Agentic Adversarial QA for Improving Domain-Specific LLMs

• Authors: Vincent Grari, Ciprian Tomoiaga, Sylvain Lamprier, Tatsunori Hashimoto, Marcin Detyniecki
• Subjects: cs.CL, cs.AI, cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18137
• Pdf URL: https://arxiv.org/pdf/2602.18137
• Copy Paste: [[2602.18137]] Agentic Adversarial QA for Improving Domain-Specific LLMs(https://arxiv.org/abs/2602.18137)
• Keywords: robust, extraction, large language model
• Abstract: Large Language Models (LLMs), despite extensive pretraining on broad internet corpora, often struggle to adapt effectively to specialized domains. There is growing interest in fine-tuning these models for such domains; however, progress is constrained by the scarcity and limited coverage of high-quality, task-relevant data. To address this, synthetic data generation methods such as paraphrasing or knowledge extraction are commonly applied. Although these approaches excel at factual recall and conceptual knowledge, they suffer from two critical shortcomings: (i) they provide minimal support for interpretive reasoning capabilities in these specialized domains, and (ii) they often produce synthetic corpora that are excessively large and redundant, resulting in poor sample efficiency. To overcome these gaps, we propose an adversarial question-generation framework that produces a compact set of semantically challenging questions. These questions are constructed by comparing the outputs of the model to be adapted and a robust expert model grounded in reference documents, using an iterative, feedback-driven process designed to reveal and address comprehension gaps. Evaluation on specialized subsets of the LegalBench corpus demonstrates that our method achieves greater accuracy with substantially fewer synthetic samples.

Title: Advection-Diffusion on Graphs: A Bakry-Emery Laplacian for Spectral Graph Neural Networks

• Authors: Pierre-Gabriel Berlureau, Ali Hariri, Victor Kawasaki-Borruat, Mia Zosso, Pierre Vandergheynst
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18141
• Pdf URL: https://arxiv.org/pdf/2602.18141
• Copy Paste: [[2602.18141]] Advection-Diffusion on Graphs: A Bakry-Emery Laplacian for Spectral Graph Neural Networks(https://arxiv.org/abs/2602.18141)
• Keywords: diffusion, transformer
• Abstract: Graph Neural Networks (GNNs) often struggle to propagate information across long distances due to oversmoothing and oversquashing. Existing remedies such as graph transformers or rewiring typically incur high computational cost or require altering the graph structure. We introduce a Bakry-Emery graph Laplacian that integrates diffusion and advection through a learnable node-wise potential, inducing task-dependent propagation dynamics without modifying topology. This operator has a well-behaved spectral decomposition and acts as a drop-in replacement for standard Laplacians in spectral GNNs. Building on this insight, we develop mu-ChebNet, a spectral architecture that jointly learns the potential and Chebyshev filters, effectively bridging message-passing adaptivity and spectral efficiency. Our theoretical analysis shows how the potential modulates the spectrum, enabling control of key graph properties. Empirically, mu-ChebNet delivers consistent gains on synthetic long-range reasoning tasks, as well as real-world benchmarks, while offering an interpretable routing field that reveals how information flows through the graph. This establishes the Bakry-Emery Laplacian as a principled and efficient foundation for adaptive spectral graph learning.

Title: Detecting Contextual Hallucinations in LLMs with Frequency-Aware Attention

• Authors: Siya Qi, Yudong Chen, Runcong Zhao, Qinglin Zhu, Zhanghao Hu, Wei Liu, Yulan He, Zheng Yuan, Lin Gui
• Subjects: cs.CL
• Abstract URL: https://arxiv.org/abs/2602.18145
• Pdf URL: https://arxiv.org/pdf/2602.18145
• Copy Paste: [[2602.18145]] Detecting Contextual Hallucinations in LLMs with Frequency-Aware Attention(https://arxiv.org/abs/2602.18145)
• Keywords: large language model
• Abstract: Hallucination detection is critical for ensuring the reliability of large language models (LLMs) in context-based generation. Prior work has explored intrinsic signals available during generation, among which attention offers a direct view of grounding behavior. However, existing approaches typically rely on coarse summaries that fail to capture fine-grained instabilities in attention. Inspired by signal processing, we introduce a frequency-aware perspective on attention by analyzing its variation during generation. We model attention distributions as discrete signals and extract high-frequency components that reflect rapid local changes in attention. Our analysis reveals that hallucinated tokens are associated with high-frequency attention energy, reflecting fragmented and unstable grounding behavior. Based on this insight, we develop a lightweight hallucination detector using high-frequency attention features. Experiments on the RAGTruth and HalluRAG benchmarks show that our approach achieves performance gains over verification-based, internal-representation-based, and attention-based methods across models and tasks.

Title: The Statistical Signature of LLMs

• Authors: Ortal Hadad, Edoardo Loru, Jacopo Nudo, Niccolò Di Marco, Matteo Cinelli, Walter Quattrociocchi
• Subjects: cs.CL, cs.CY, physics.soc-ph
• Abstract URL: https://arxiv.org/abs/2602.18152
• Pdf URL: https://arxiv.org/pdf/2602.18152
• Copy Paste: [[2602.18152]] The Statistical Signature of LLMs(https://arxiv.org/abs/2602.18152)
• Keywords: robust, generative, large language model
• Abstract: Large language models generate text through probabilistic sampling from high-dimensional distributions, yet how this process reshapes the structural statistical organization of language remains incompletely characterized. Here we show that lossless compression provides a simple, model-agnostic measure of statistical regularity that differentiates generative regimes directly from surface text. We analyze compression behavior across three progressively more complex information ecosystems: controlled human-LLM continuations, generative mediation of a knowledge infrastructure (Wikipedia vs. Grokipedia), and fully synthetic social interaction environments (Moltbook vs. Reddit). Across settings, compression reveals a persistent structural signature of probabilistic generation. In controlled and mediated contexts, LLM-produced language exhibits higher structural regularity and compressibility than human-written text, consistent with a concentration of output within highly recurrent statistical patterns. However, this signature shows scale dependence: in fragmented interaction environments the separation attenuates, suggesting a fundamental limit to surface-level distinguishability at small scales. This compressibility-based separation emerges consistently across models, tasks, and domains and can be observed directly from surface text without relying on model internals or semantic evaluation. Overall, our findings introduce a simple and robust framework for quantifying how generative systems reshape textual production, offering a structural perspective on the evolving complexity of communication.

Title: FENCE: A Financial and Multimodal Jailbreak Detection Dataset

• Authors: Mirae Kim, Seonghun Jeong, Youngjun Kwak
• Subjects: cs.CL, cs.AI, cs.DB
• Abstract URL: https://arxiv.org/abs/2602.18154
• Pdf URL: https://arxiv.org/pdf/2602.18154
• Copy Paste: [[2602.18154]] FENCE: A Financial and Multimodal Jailbreak Detection Dataset(https://arxiv.org/abs/2602.18154)
• Keywords: attack, robust, large language model
• Abstract: Jailbreaking poses a significant risk to the deployment of Large Language Models (LLMs) and Vision Language Models (VLMs). VLMs are particularly vulnerable because they process both text and images, creating broader attack surfaces. However, available resources for jailbreak detection are scarce, particularly in finance. To address this gap, we present FENCE, a bilingual (Korean-English) multimodal dataset for training and evaluating jailbreak detectors in financial applications. FENCE emphasizes domain realism through finance-relevant queries paired with image-grounded threats. Experiments with commercial and open-source VLMs reveal consistent vulnerabilities, with GPT-4o showing measurable attack success rates and open-source models displaying greater exposure. A baseline detector trained on FENCE achieves 99 percent in-distribution accuracy and maintains strong performance on external benchmarks, underscoring the dataset's robustness for training reliable detection models. FENCE provides a focused resource for advancing multimodal jailbreak detection in finance and for supporting safer, more reliable AI systems in sensitive domains. Warning: This paper includes example data that may be offensive.

Title: Unifying Formal Explanations: A Complexity-Theoretic Perspective

• Authors: Shahaf Bassan, Xuanxiang Huang, Guy Katz
• Subjects: cs.LG, cs.CC, cs.DS, math.OC
• Abstract URL: https://arxiv.org/abs/2602.18160
• Pdf URL: https://arxiv.org/pdf/2602.18160
• Copy Paste: [[2602.18160]] Unifying Formal Explanations: A Complexity-Theoretic Perspective(https://arxiv.org/abs/2602.18160)
• Keywords: interpretability, explainability
• Abstract: Previous work has explored the computational complexity of deriving two fundamental types of explanations for ML model predictions: (1) *sufficient reasons*, which are subsets of input features that, when fixed, determine a prediction, and (2) *contrastive reasons*, which are subsets of input features that, when modified, alter a prediction. Prior studies have examined these explanations in different contexts, such as non-probabilistic versus probabilistic frameworks and local versus global settings. In this study, we introduce a unified framework for analyzing these explanations, demonstrating that they can all be characterized through the minimization of a unified probabilistic value function. We then prove that the complexity of these computations is influenced by three key properties of the value function: (1) *monotonicity*, (2) *submodularity*, and (3) *supermodularity* - which are three fundamental properties in *combinatorial optimization*. Our findings uncover some counterintuitive results regarding the nature of these properties within the explanation settings examined. For instance, although the *local* value functions do not exhibit monotonicity or submodularity/supermodularity whatsoever, we demonstrate that the *global* value functions do possess these properties. This distinction enables us to prove a series of novel polynomial-time results for computing various explanations with provable guarantees in the global explainability setting, across a range of ML models that span the interpretability spectrum, such as neural networks, decision trees, and tree ensembles. In contrast, we show that even highly simplified versions of these explanations become NP-hard to compute in the corresponding local explainability setting.

Title: A Deep Surrogate Model for Robust and Generalizable Long-Term Blast Wave Prediction

• Authors: Danning Jing, Xinhai Chen, Xifeng Pu, Jie Hu, Chao Huang, Xuguang Chen, Qinglin Wang, Jie Liu
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18168
• Pdf URL: https://arxiv.org/pdf/2602.18168
• Copy Paste: [[2602.18168]] A Deep Surrogate Model for Robust and Generalizable Long-Term Blast Wave Prediction(https://arxiv.org/abs/2602.18168)
• Keywords: robust
• Abstract: Accurately modeling the spatio-temporal dynamics of blast wave propagation remains a longstanding challenge due to its highly nonlinear behavior, sharp gradients, and burdensome computational cost. While machine learning-based surrogate models offer fast inference as a promising alternative, they suffer from degraded accuracy, particularly evaluated on complex urban layouts or out-of-distribution scenarios. Moreover, autoregressive prediction strategies in such models are prone to error accumulation over long forecasting horizons, limiting their robustness for extended-time simulations. To address these limitations, we propose RGD-Blast, a robust and generalizable deep surrogate model for high-fidelity, long-term blast wave forecasting. RGD-Blast incorporates a multi-scale module to capture both global flow patterns and local boundary interactions, effectively mitigating error accumulation during autoregressive prediction. We introduce a dynamic-static feature coupling mechanism that fuses time-varying pressure fields with static source and layout features, thereby enhancing out-of-distribution generalization. Experiments demonstrate that RGD-Blast achieves a two-order-of-magnitude speedup over traditional numerical methods while maintaining comparable accuracy. In generalization tests on unseen building layouts, the model achieves an average RMSE below 0.01 and an R2 exceeding 0.89 over 280 consecutive time steps. Additional evaluations under varying blast source locations and explosive charge weights further validate its generalization, substantially advancing the state of the art in long-term blast wave modeling.

Title: Click it or Leave it: Detecting and Spoiling Clickbait with Informativeness Measures and Large Language Models

• Authors: Wojciech Michaluk, Tymoteusz Urban, Mateusz Kubita, Soveatin Kuntur, Anna Wroblewska
• Subjects: cs.CL, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18171
• Pdf URL: https://arxiv.org/pdf/2602.18171
• Copy Paste: [[2602.18171]] Click it or Leave it: Detecting and Spoiling Clickbait with Informativeness Measures and Large Language Models(https://arxiv.org/abs/2602.18171)
• Keywords: interpretability, transformer, large language model
• Abstract: Clickbait headlines degrade the quality of online information and undermine user trust. We present a hybrid approach to clickbait detection that combines transformer-based text embeddings with linguistically motivated informativeness features. Using natural language processing techniques, we evaluate classical vectorizers, word embedding baselines, and large language model embeddings paired with tree-based classifiers. Our best-performing model, XGBoost over embeddings augmented with 15 explicit features, achieves an F1-score of 91\%, outperforming TF-IDF, Word2Vec, GloVe, LLM prompt based classification, and feature-only baselines. The proposed feature set enhances interpretability by highlighting salient linguistic cues such as second-person pronouns, superlatives, numerals, and attention-oriented punctuation, enabling transparent and well-calibrated clickbait predictions. We release code and trained models to support reproducible research.

Title: Can AI Lower the Barrier to Cybersecurity? A Human-Centered Mixed-Methods Study of Novice CTF Learning

• Authors: Cathrin Schachner, Jasmin Wachter
• Subjects: cs.CR, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18172
• Pdf URL: https://arxiv.org/pdf/2602.18172
• Copy Paste: [[2602.18172]] Can AI Lower the Barrier to Cybersecurity? A Human-Centered Mixed-Methods Study of Novice CTF Learning(https://arxiv.org/abs/2602.18172)
• Keywords: security
• Abstract: Capture-the-Flag (CTF) competitions serve as gateways into offensive cybersecurity, yet they often present steep barriers for novices due to complex toolchains and opaque workflows. Recently, agentic AI frameworks for cybersecurity promise to lower these barriers by automating and coordinating penetration testing tasks. However, their role in shaping novice learning remains underexplored. We present a human-centered, mixed-methods case study examining how agentic AI frameworks -- here Cybersecurity AI (CAI) -- mediates novice entry into CTF-based penetration testing. An undergraduate student without prior hacking experience attempted to approach performance benchmarks from a national cybersecurity challenge using CAI. Quantitative performance metrics were complemented by structured reflective analysis of learning progression and AI interaction patterns. Our thematic analysis suggest that agentic AI reduces initial entry barriers by providing overview, structure and guidance, thereby lowering the cognitive workload during early engagement. Quantitatively, the observed extensive exploration of strategies and low per-strategy execution time potetially facilitatates cybersecurity training on meta, i.e. strategic levels. At the same time, AI-assisted cybersecurity education introduces new challenges related to trust, dependency, and responsible use. We discuss implications for human-centered AI-supported cybersecurity education and outline open questions for future research.

Title: Improving Sampling for Masked Diffusion Models via Information Gain

• Authors: Kaisen Yang, Jayden Teoh, Kaicheng Yang, Yitong Zhang, Alex Lamb
• Subjects: cs.CL
• Abstract URL: https://arxiv.org/abs/2602.18176
• Pdf URL: https://arxiv.org/pdf/2602.18176
• Copy Paste: [[2602.18176]] Improving Sampling for Masked Diffusion Models via Information Gain(https://arxiv.org/abs/2602.18176)
• Keywords: diffusion
• Abstract: Masked Diffusion Models (MDMs) offer greater flexibility in decoding order than autoregressive models but require careful planning to achieve high-quality generation. Existing samplers typically adopt greedy heuristics, prioritizing positions with the highest local certainty to decode at each step. Through failure case analysis, we identify a fundamental limitation of this approach: it neglects the downstream impact of current decoding choices on subsequent steps and fails to minimize cumulative uncertainty. In particular, these methods do not fully exploit the non-causal nature of MDMs, which enables evaluating how a decoding decision reshapes token probabilities/uncertainty across all remaining masked positions. To bridge this gap, we propose the Info-Gain Sampler, a principled decoding framework that balances immediate uncertainty with information gain over future masked tokens. Extensive evaluations across diverse architectures and tasks (reasoning, coding, creative writing, and image generation) demonstrate that Info-Gain Sampler consistently outperforms existing samplers for MDMs. For instance, it achieves a 3.6% improvement in average accuracy on reasoning tasks and a 63.1% win-rate in creative writing. Notably, on reasoning tasks it reduces cumulative uncertainty from 78.4 to 48.6, outperforming the best baseline by a large margin. The code will be available at this https URL.

Title: Evaluating Graphical Perception Capabilities of Vision Transformers

• Authors: Poonam Poonam, Pere-Pau Vázquez, Timo Ropinski
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18178
• Pdf URL: https://arxiv.org/pdf/2602.18178
• Copy Paste: [[2602.18178]] Evaluating Graphical Perception Capabilities of Vision Transformers(https://arxiv.org/abs/2602.18178)
• Keywords: transformer
• Abstract: Vision Transformers, ViTs, have emerged as a powerful alternative to convolutional neural networks, CNNs, in a variety of image-based tasks. While CNNs have previously been evaluated for their ability to perform graphical perception tasks, which are essential for interpreting visualizations, the perceptual capabilities of ViTs remain largely unexplored. In this work, we investigate the performance of ViTs in elementary visual judgment tasks inspired by the foundational studies of Cleveland and McGill, which quantified the accuracy of human perception across different visual encodings. Inspired by their study, we benchmark ViTs against CNNs and human participants in a series of controlled graphical perception tasks. Our results reveal that, although ViTs demonstrate strong performance in general vision tasks, their alignment with human-like graphical perception in the visualization domain is limited. This study highlights key perceptual gaps and points to important considerations for the application of ViTs in visualization systems and graphical perceptual modeling.

Title: BLM-Guard: Explainable Multimodal Ad Moderation with Chain-of-Thought and Policy-Aligned Rewards

• Authors: Yiran Yang, Zhaowei Liu, Yuan Yuan, Yukun Song, Xiong Ma, Yinghao Song, Xiangji Zeng, Lu Sun, Yulu Wang, Hai Zhou, Shuai Cui, Zhaohan Gong, Jiefei Zhang
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18193
• Pdf URL: https://arxiv.org/pdf/2602.18193
• Copy Paste: [[2602.18193]] BLM-Guard: Explainable Multimodal Ad Moderation with Chain-of-Thought and Policy-Aligned Rewards(https://arxiv.org/abs/2602.18193)
• Keywords: robust
• Abstract: Short-video platforms now host vast multimodal ads whose deceptive visuals, speech and subtitles demand finer-grained, policy-driven moderation than community safety filters. We present BLM-Guard, a content-audit framework for commercial ads that fuses Chain-of-Thought reasoning with rule-based policy principles and a critic-guided reward. A rule-driven ICoT data-synthesis pipeline jump-starts training by generating structured scene descriptions, reasoning chains and labels, cutting annotation costs. Reinforcement learning then refines the model using a composite reward balancing causal coherence with policy adherence. A multitask architecture models intra-modal manipulations (e.g., exaggerated imagery) and cross-modal mismatches (e.g., subtitle-speech drift), boosting robustness. Experiments on real short-video ads show BLM-Guard surpasses strong baselines in accuracy, consistency and generalization.

Title: LERD: Latent Event-Relational Dynamics for Neurodegenerative Classification

• Authors: Hairong Chen, Yicheng Feng, Ziyu Jia, Samir Bhatt, Hengguan Huang
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18195
• Pdf URL: https://arxiv.org/pdf/2602.18195
• Copy Paste: [[2602.18195]] LERD: Latent Event-Relational Dynamics for Neurodegenerative Classification(https://arxiv.org/abs/2602.18195)
• Keywords: generative
• Abstract: Alzheimer's disease (AD) alters brain electrophysiology and disrupts multichannel EEG dynamics, making accurate and clinically useful EEG-based diagnosis increasingly important for screening and disease monitoring. However, many existing approaches rely on black-box classifiers and do not explicitly model the underlying dynamics that generate observed signals. To address these limitations, we propose LERD, an end-to-end Bayesian electrophysiological neural dynamical system that infers latent neural events and their relational structure directly from multichannel EEG without event or interaction annotations. LERD combines a continuous-time event inference module with a stochastic event-generation process to capture flexible temporal patterns, while incorporating an electrophysiology-inspired dynamical prior to guide learning in a principled way. We further provide theoretical analysis that yields a tractable bound for training and stability guarantees for the inferred relational dynamics. Extensive experiments on synthetic benchmarks and two real-world AD EEG cohorts demonstrate that LERD consistently outperforms strong baselines and yields physiology-aligned latent summaries that help characterize group-level dynamical differences.

Title: Generative Model via Quantile Assignment

• Authors: Georgi Hrusanov, Oliver Y. Chén, Julien S. Bodelet
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18216
• Pdf URL: https://arxiv.org/pdf/2602.18216
• Copy Paste: [[2602.18216]] Generative Model via Quantile Assignment(https://arxiv.org/abs/2602.18216)
• Keywords: robust, diffusion, generative
• Abstract: Deep Generative models (DGMs) play two key roles in modern machine learning: (i) producing new information (e.g., image synthesis) and (ii) reducing dimensionality. However, traditional architectures often rely on auxiliary networks such as encoders in Variational Autoencoders (VAEs) or discriminators in Generative Adversarial Networks (GANs), which introduce training instability, computational overhead, and risks like mode collapse. We present NeuroSQL, a new generative paradigm that eliminates the need for auxiliary networks by learning low-dimensional latent representations implicitly. NeuroSQL leverages an asymptotic approximation that expresses the latent variables as the solution to an optimal transportation problem. Specifically, NeuroSQL learns the latent variables by solving a linear assignment problem and then passes the latent information to a standalone generator. We benchmark its performance against GANs, VAEs, and a budget-matched diffusion baseline on four datasets: handwritten digits (MNIST), faces (CelebA), animal faces (AFHQ), and brain images (OASIS). Compared to VAEs, GANs, and diffusion models: (1) in terms of image quality, NeuroSQL achieves overall lower mean pixel distance between synthetic and authentic images and stronger perceptual/structural fidelity; (2) computationally, NeuroSQL requires the least training time; and (3) practically, NeuroSQL provides an effective solution for generating synthetic data with limited training samples. By embracing quantile assignment rather than an encoder, NeuroSQL provides a fast, stable, and robust way to generate synthetic data with minimal information loss.

Title: [Re] Benchmarking LLM Capabilities in Negotiation through Scoreable Games

• Authors: Jorge Carrasco Pollo, Ioannis Kapetangeorgis, Joshua Rosenthal, John Hua Yao
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18230
• Pdf URL: https://arxiv.org/pdf/2602.18230
• Copy Paste: [[2602.18230]] [Re] Benchmarking LLM Capabilities in Negotiation through Scoreable Games(https://arxiv.org/abs/2602.18230)
• Keywords: robust, large language model
• Abstract: Large Language Models (LLMs) demonstrate significant potential in multi-agent negotiation tasks, yet evaluation in this domain remains challenging due to a lack of robust and generalizable benchmarks. Abdelnabi et al. (2024) introduce a negotiation benchmark based on Scoreable Games, with the aim of developing a highly complex and realistic evaluation framework for LLMs. Our work investigates the reproducibility of claims in their benchmark, and provides a deeper understanding of its usability and generalizability. We replicate the original experiments on additional models, and introduce additional metrics to verify negotiation quality and evenness of evaluation. Our findings reveal that while the benchmark is indeed complex, model comparison is ambiguous, raising questions about its objectivity. Furthermore, we identify limitations in the experimental setup, particularly in information leakage detection and thoroughness of the ablation study. By examining and analyzing the behavior of a wider range of models on an extended version of the benchmark, we reveal insights that provide additional context to potential users. Our results highlight the importance of context in model-comparative evaluations.

Title: Thinking by Subtraction: Confidence-Driven Contrastive Decoding for LLM Reasoning

• Authors: Lexiang Tang, Weihao Gao, Bingchen Zhao, Lu Ma, Qiao jin, Bang Yang, Yuexian Zou
• Subjects: cs.CL, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18232
• Pdf URL: https://arxiv.org/pdf/2602.18232
• Copy Paste: [[2602.18232]] Thinking by Subtraction: Confidence-Driven Contrastive Decoding for LLM Reasoning(https://arxiv.org/abs/2602.18232)
• Keywords: large language model
• Abstract: Recent work on test-time scaling for large language model (LLM) reasoning typically assumes that allocating more inference-time computation uniformly improves correctness. However, prior studies show that reasoning uncertainty is highly localized: a small subset of low-confidence tokens disproportionately contributes to reasoning errors and unnecessary output expansion. Motivated by this observation, we propose Thinking by Subtraction, a confidence-driven contrastive decoding approach that improves reasoning reliability through targeted token-level intervention. Our method, Confidence-Driven Contrastive Decoding, detects low-confidence tokens during decoding and intervenes selectively at these positions. It constructs a contrastive reference by replacing high-confidence tokens with minimal placeholders, and refines predictions by subtracting this reference distribution at low-confidence locations. Experiments show that CCD significantly improves accuracy across mathematical reasoning benchmarks while substantially reducing output length, with minimal KV-cache overhead. As a training-free method, CCD enhances reasoning reliability through targeted low-confidence intervention without computational redundancy. Our code will be made available at: this https URL.

Title: On the Adversarial Robustness of Discrete Image Tokenizers

• Authors: Rishika Bhagwatkar, Irina Rish, Nicolas Flammarion, Francesco Croce
• Subjects: cs.CV, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18252
• Pdf URL: https://arxiv.org/pdf/2602.18252
• Copy Paste: [[2602.18252]] On the Adversarial Robustness of Discrete Image Tokenizers(https://arxiv.org/abs/2602.18252)
• Keywords: attack, robust
• Abstract: Discrete image tokenizers encode visual inputs as sequences of tokens from a finite vocabulary and are gaining popularity in multimodal systems, including encoder-only, encoder-decoder, and decoder-only models. However, unlike CLIP encoders, their vulnerability to adversarial attacks has not been explored. Ours being the first work studying this topic, we first formulate attacks that aim to perturb the features extracted by discrete tokenizers, and thus change the extracted tokens. These attacks are computationally efficient, application-agnostic, and effective across classification, multimodal retrieval, and captioning tasks. Second, to defend against this vulnerability, inspired by recent work on robust CLIP encoders, we fine-tune popular tokenizers with unsupervised adversarial training, keeping all other components frozen. While unsupervised and task-agnostic, our approach significantly improves robustness to both unsupervised and end-to-end supervised attacks and generalizes well to unseen tasks and data. Unlike supervised adversarial training, our approach can leverage unlabeled images, making it more versatile. Overall, our work highlights the critical role of tokenizer robustness in downstream tasks and presents an important step in the development of safe multimodal foundation models.

Title: Simplifying Outcomes of Language Model Component Analyses with ELIA

• Authors: Aaron Louis Eidt, Nils Feldhus
• Subjects: cs.CL, cs.AI, cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18262
• Pdf URL: https://arxiv.org/pdf/2602.18262
• Copy Paste: [[2602.18262]] Simplifying Outcomes of Language Model Component Analyses with ELIA(https://arxiv.org/abs/2602.18262)
• Keywords: interpretability, large language model
• Abstract: While mechanistic interpretability has developed powerful tools to analyze the internal workings of Large Language Models (LLMs), their complexity has created an accessibility gap, limiting their use to specialists. We address this challenge by designing, building, and evaluating ELIA (Explainable Language Interpretability Analysis), an interactive web application that simplifies the outcomes of various language model component analyses for a broader audience. The system integrates three key techniques -- Attribution Analysis, Function Vector Analysis, and Circuit Tracing -- and introduces a novel methodology: using a vision-language model to automatically generate natural language explanations (NLEs) for the complex visualizations produced by these methods. The effectiveness of this approach was empirically validated through a mixed-methods user study, which revealed a clear preference for interactive, explorable interfaces over simpler, static visualizations. A key finding was that the AI-powered explanations helped bridge the knowledge gap for non-experts; a statistical analysis showed no significant correlation between a user's prior LLM experience and their comprehension scores, suggesting that the system reduced barriers to comprehension across experience levels. We conclude that an AI system can indeed simplify complex model analyses, but its true power is unlocked when paired with thoughtful, user-centered design that prioritizes interactivity, specificity, and narrative guidance.

Title: Many Tools, Few Exploitable Vulnerabilities: A Survey of 246 Static Code Analyzers for Security

• Authors: Kevin Hermann, Sven Peldszus, Thorsten Berger
• Subjects: cs.CR, cs.SE
• Abstract URL: https://arxiv.org/abs/2602.18270
• Pdf URL: https://arxiv.org/pdf/2602.18270
• Copy Paste: [[2602.18270]] Many Tools, Few Exploitable Vulnerabilities: A Survey of 246 Static Code Analyzers for Security(https://arxiv.org/abs/2602.18270)
• Keywords: security, robust
• Abstract: Static security analysis is a widely used technique for detecting software vulnerabilities across a wide range of weaknesses, application domains, and programming languages. While prior work surveyed static analyzes for specific weaknesses or application domains, no overview of the entire security landscape exists. We present a systematic literature review of 246 static security analyzers concerning their targeted vulnerabilities, application domains, analysis techniques, evaluation methods, and limitations. We observe that most analyzers focus on a limited set of weaknesses, that the vulnerabilities they detect are rarely exploitable, and that evaluations use custom benchmarks that are too small to enable robust assessment.

Title: DEIG: Detail-Enhanced Instance Generation with Fine-Grained Semantic Control

• Authors: Shiyan Du, Conghan Yue, Xinyu Cheng, Dongyu Zhang
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18282
• Pdf URL: https://arxiv.org/pdf/2602.18282
• Copy Paste: [[2602.18282]] DEIG: Detail-Enhanced Instance Generation with Fine-Grained Semantic Control(https://arxiv.org/abs/2602.18282)
• Keywords: diffusion
• Abstract: Multi-Instance Generation has advanced significantly in spatial placement and attribute binding. However, existing approaches still face challenges in fine-grained semantic understanding, particularly when dealing with complex textual descriptions. To overcome these limitations, we propose DEIG, a novel framework for fine-grained and controllable multi-instance generation. DEIG integrates an Instance Detail Extractor (IDE) that transforms text encoder embeddings into compact, instance-aware representations, and a Detail Fusion Module (DFM) that applies instance-based masked attention to prevent attribute leakage across instances. These components enable DEIG to generate visually coherent multi-instance scenes that precisely match rich, localized textual descriptions. To support fine-grained supervision, we construct a high-quality dataset with detailed, compositional instance captions generated by VLMs. We also introduce DEIG-Bench, a new benchmark with region-level annotations and multi-attribute prompts for both humans and objects. Experiments demonstrate that DEIG consistently outperforms existing approaches across multiple benchmarks in spatial consistency, semantic accuracy, and compositional generalization. Moreover, DEIG functions as a plug-and-play module, making it easily integrable into standard diffusion-based pipelines.

Title: Detecting PowerShell-based Fileless Cryptojacking Attacks Using Machine Learning

• Authors: Said Varlioglu, Nelly Elsayed, Murat Ozer, Zag ElSayed, John M. Emmert
• Subjects: cs.CR
• Abstract URL: https://arxiv.org/abs/2602.18285
• Pdf URL: https://arxiv.org/pdf/2602.18285
• Copy Paste: [[2602.18285]] Detecting PowerShell-based Fileless Cryptojacking Attacks Using Machine Learning(https://arxiv.org/abs/2602.18285)
• Keywords: attack, steal
• Abstract: With the emergence of remote code execution (RCE) vulnerabilities in ubiquitous libraries and advanced social engineering techniques, threat actors have started conducting widespread fileless cryptojacking attacks. These attacks have become effective with stealthy techniques based on PowerShell-based exploitation in Windows OS environments. Even if attacks are detected and malicious scripts removed, processes may remain operational on victim endpoints, creating a significant challenge for detection mechanisms. In this paper, we conducted an experimental study with a collected dataset on detecting PowerShell-based fileless cryptojacking scripts. The results showed that Abstract Syntax Tree (AST)-based fine-tuned CodeBERT achieved a high recall rate, proving the importance of the use of AST integration and fine-tuned pre-trained models for programming language.

Title: On the Semantic and Syntactic Information Encoded in Proto-Tokens for One-Step Text Reconstruction

• Authors: Ivan Bondarenko, Egor Palkin, Fedor Tikunov
• Subjects: cs.LG, cs.CL
• Abstract URL: https://arxiv.org/abs/2602.18301
• Pdf URL: https://arxiv.org/pdf/2602.18301
• Copy Paste: [[2602.18301]] On the Semantic and Syntactic Information Encoded in Proto-Tokens for One-Step Text Reconstruction(https://arxiv.org/abs/2602.18301)
• Keywords: large language model
• Abstract: Autoregressive large language models (LLMs) generate text token-by-token, requiring n forward passes to produce a sequence of length n. Recent work, Exploring the Latent Capacity of LLMs for One-Step Text Reconstruction (Mezentsev and Oseledets), shows that frozen LLMs can reconstruct hundreds of tokens from only two learned proto-tokens in a single forward pass, suggesting a path beyond the autoregressive paradigm. In this paper, we study what information these proto-tokens encode and how they behave under reconstruction and controlled constraints. We perform a series of experiments aimed at disentangling semantic and syntactic content in the two proto-tokens, analyzing stability properties of the e-token, and visualizing attention patterns to the e-token during reconstruction. Finally, we test two regularization schemes for "imposing" semantic structure on the e-token using teacher embeddings, including an anchor-based loss and a relational distillation objective. Our results indicate that the m-token tends to capture semantic information more strongly than the e-token under standard optimization; anchor-based constraints trade off sharply with reconstruction accuracy; and relational distillation can transfer batch-level semantic relations into the proto-token space without sacrificing reconstruction quality, supporting the feasibility of future non-autoregressive seq2seq systems that predict proto-tokens as an intermediate representation.

Title: FeatureBleed: Inferring Private Enriched Attributes From Sparsity-Optimized AI Accelerators

• Authors: Darsh Asher, Farshad Dizani, Joshua Kalyanapu, Rosario Cammarota, Aydin Aysu, Samira Mirbagher Ajorpaz
• Subjects: cs.CR
• Abstract URL: https://arxiv.org/abs/2602.18304
• Pdf URL: https://arxiv.org/pdf/2602.18304
• Copy Paste: [[2602.18304]] FeatureBleed: Inferring Private Enriched Attributes From Sparsity-Optimized AI Accelerators(https://arxiv.org/abs/2602.18304)
• Keywords: privacy, protect, defense, attack, steal
• Abstract: Backend enrichment is now widely deployed in sensitive domains such as product recommendation pipelines, healthcare, and finance, where models are trained on confidential data and retrieve private features whose values influence inference behavior while remaining hidden from the API caller. This paper presents the first hardware-level backend retrieval data-stealing attack, showing that accelerator optimizations designed for performance can directly undermine data confidentiality and bypass state-of-the-art privacy defenses. Our attack, FEATUREBLEED, exploits zero-skipping in AI accelerators to infer private backend-retrieved features solely through end-to-end timing, without relying on power analysis, DVFS manipulation, or shared-cache side channels. We evaluate FEATUREBLEED on three datasets spanning medical and non-medical domains: Texas-100X (clinical records), OrganAMNIST (medical imaging), and Census-19 (socioeconomic data). We further evaluate FEATUREBLEED across three hardware backends (Intel AVX, Intel AMX, and NVIDIA A100) and three model architectures (DNNs, CNNs, and hybrid CNN-MLP pipelines), demonstrating that the leakage generalizes across CPU and GPU accelerators, data modalities, and application domains, with an adversarial advantage of up to 98.87 percentage points. Finally, we identify the root cause of the leakage as sparsity-driven zero-skipping in modern hardware. We quantify the privacy-performance-power trade-off: disabling zero-skipping increases Intel AMX per-operation energy by up to 25 percent and incurs 100 percent performance overhead. We propose a padding-based defense that masks timing leakage by equalizing responses to the worst-case execution time, achieving protection with only 7.24 percent average performance overhead and no additional power cost.

Title: Multi-Level Conditioning by Pairing Localized Text and Sketch for Fashion Image Generation

• Authors: Ziyue Liu, Davide Talon, Federico Girella, Zanxi Ruan, Mattia Mondo, Loris Bazzani, Yiming Wang, Marco Cristani
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18309
• Pdf URL: https://arxiv.org/pdf/2602.18309
• Copy Paste: [[2602.18309]] Multi-Level Conditioning by Pairing Localized Text and Sketch for Fashion Image Generation(https://arxiv.org/abs/2602.18309)
• Keywords: robust, diffusion
• Abstract: Sketches offer designers a concise yet expressive medium for early-stage fashion ideation by specifying structure, silhouette, and spatial relationships, while textual descriptions complement sketches to convey material, color, and stylistic details. Effectively combining textual and visual modalities requires adherence to the sketch visual structure when leveraging the guidance of localized attributes from text. We present LOcalized Text and Sketch with multi-level guidance (LOTS), a framework that enhances fashion image generation by combining global sketch guidance with multiple localized sketch-text pairs. LOTS employs a Multi-level Conditioning Stage to independently encode local features within a shared latent space while maintaining global structural coordination. Then, the Diffusion Pair Guidance stage integrates both local and global conditioning via attention-based guidance within the diffusion model's multi-step denoising process. To validate our method, we develop Sketchy, the first fashion dataset where multiple text-sketch pairs are provided per image. Sketchy provides high-quality, clean sketches with a professional look and consistent structure. To assess robustness beyond this setting, we also include an "in the wild" split with non-expert sketches, featuring higher variability and imperfections. Experiments demonstrate that our method strengthens global structural adherence while leveraging richer localized semantic guidance, achieving improvement over state-of-the-art. The dataset, platform, and code are publicly available.

Title: Diff2DGS: Reliable Reconstruction of Occluded Surgical Scenes via 2D Gaussian Splatting

• Authors: Tianyi Song, Danail Stoyanov, Evangelos Mazomenos, Francisco Vasconcelos
• Subjects: cs.CV, cs.GR, cs.RO
• Abstract URL: https://arxiv.org/abs/2602.18314
• Pdf URL: https://arxiv.org/pdf/2602.18314
• Copy Paste: [[2602.18314]] Diff2DGS: Reliable Reconstruction of Occluded Surgical Scenes via 2D Gaussian Splatting(https://arxiv.org/abs/2602.18314)
• Keywords: diffusion
• Abstract: Real-time reconstruction of deformable surgical scenes is vital for advancing robotic surgery, improving surgeon guidance, and enabling automation. Recent methods achieve dense reconstructions from da Vinci robotic surgery videos, with Gaussian Splatting (GS) offering real-time performance via graphics acceleration. However, reconstruction quality in occluded regions remains limited, and depth accuracy has not been fully assessed, as benchmarks like EndoNeRF and StereoMIS lack 3D ground truth. We propose Diff2DGS, a novel two-stage framework for reliable 3D reconstruction of occluded surgical scenes. In the first stage, a diffusion-based video module with temporal priors inpaints tissue occluded by instruments with high spatial-temporal consistency. In the second stage, we adapt 2D Gaussian Splatting (2DGS) with a Learnable Deformation Model (LDM) to capture dynamic tissue deformation and anatomical geometry. We also extend evaluation beyond prior image-quality metrics by performing quantitative depth accuracy analysis on the SCARED dataset. Diff2DGS outperforms state-of-the-art approaches in both appearance and geometry, reaching 38.02 dB PSNR on EndoNeRF and 34.40 dB on StereoMIS. Furthermore, our experiments demonstrate that optimizing for image quality alone does not necessarily translate into optimal 3D reconstruction accuracy. To address this, we further optimize the depth quality of the reconstructed 3D results, ensuring more faithful geometry in addition to high-fidelity appearance.

Title: Unifying Color and Lightness Correction with View-Adaptive Curve Adjustment for Robust 3D Novel View Synthesis

• Authors: Ziteng Cui, Shuhong Liu, Xiaoyu Dong, Xuangeng Chu, Lin Gu, Ming-Hsuan Yang, Tatsuya Harada
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18322
• Pdf URL: https://arxiv.org/pdf/2602.18322
• Copy Paste: [[2602.18322]] Unifying Color and Lightness Correction with View-Adaptive Curve Adjustment for Robust 3D Novel View Synthesis(https://arxiv.org/abs/2602.18322)
• Keywords: robust
• Abstract: High-quality image acquisition in real-world environments remains challenging due to complex illumination variations and inherent limitations of camera imaging pipelines. These issues are exacerbated in multi-view capture, where differences in lighting, sensor responses, and image signal processor (ISP) configurations introduce photometric and chromatic inconsistencies that violate the assumptions of photometric consistency underlying modern 3D novel view synthesis (NVS) methods, including Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS), leading to degraded reconstruction and rendering quality. We propose Luminance-GS++, a 3DGS-based framework for robust NVS under diverse illumination conditions. Our method combines a globally view-adaptive lightness adjustment with a local pixel-wise residual refinement for precise color correction. We further design unsupervised objectives that jointly enforce lightness correction and multi-view geometric and photometric consistency. Extensive experiments demonstrate state-of-the-art performance across challenging scenarios, including low-light, overexposure, and complex luminance and chromatic variations. Unlike prior approaches that modify the underlying representation, our method preserves the explicit 3DGS formulation, improving reconstruction fidelity while maintaining real-time rendering efficiency.

Title: On the "Induction Bias" in Sequence Models

• Authors: M.Reza Ebrahimi, Michaël Defferrard, Sunny Panchal, Roland Memisevic
• Subjects: cs.LG, cs.CL
• Abstract URL: https://arxiv.org/abs/2602.18333
• Pdf URL: https://arxiv.org/pdf/2602.18333
• Copy Paste: [[2602.18333]] On the "Induction Bias" in Sequence Models(https://arxiv.org/abs/2602.18333)
• Keywords: transformer
• Abstract: Despite the remarkable practical success of transformer-based language models, recent work has raised concerns about their ability to perform state tracking. In particular, a growing body of literature has shown this limitation primarily through failures in out-of-distribution (OOD) generalization, such as length extrapolation. In this work, we shift attention to the in-distribution implications of these limitations. We conduct a large-scale experimental study of the data efficiency of transformers and recurrent neural networks (RNNs) across multiple supervision regimes. We find that the amount of training data required by transformers grows much more rapidly with state-space size and sequence length than for RNNs. Furthermore, we analyze the extent to which learned state-tracking mechanisms are shared across different sequence lengths. We show that transformers exhibit negligible or even detrimental weight sharing across lengths, indicating that they learn length-specific solutions in isolation. In contrast, recurrent models exhibit effective amortized learning by sharing weights across lengths, allowing data from one sequence length to improve performance on others. Together, these results demonstrate that state tracking remains a fundamental challenge for transformers, even when training and evaluation distributions match.

Title: Vichara: Appellate Judgment Prediction and Explanation for the Indian Judicial System

• Authors: Pavithra PM Nair, Preethu Rose Anish
• Subjects: cs.CL, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18346
• Pdf URL: https://arxiv.org/pdf/2602.18346
• Copy Paste: [[2602.18346]] Vichara: Appellate Judgment Prediction and Explanation for the Indian Judicial System(https://arxiv.org/abs/2602.18346)
• Keywords: interpretability, large language model
• Abstract: In jurisdictions like India, where courts face an extensive backlog of cases, artificial intelligence offers transformative potential for legal judgment prediction. A critical subset of this backlog comprises appellate cases, which are formal decisions issued by higher courts reviewing the rulings of lower courts. To this end, we present Vichara, a novel framework tailored to the Indian judicial system that predicts and explains appellate judgments. Vichara processes English-language appellate case proceeding documents and decomposes them into decision points. Decision points are discrete legal determinations that encapsulate the legal issue, deciding authority, outcome, reasoning, and temporal context. The structured representation isolates the core determinations and their context, enabling accurate predictions and interpretable explanations. Vichara's explanations follow a structured format inspired by the IRAC (Issue-Rule-Application-Conclusion) framework and adapted for Indian legal reasoning. This enhances interpretability, allowing legal professionals to assess the soundness of predictions efficiently. We evaluate Vichara on two datasets, PredEx and the expert-annotated subset of the Indian Legal Documents Corpus (ILDC_expert), using four large language models: GPT-4o mini, Llama-3.1-8B, Mistral-7B, and Qwen2.5-7B. Vichara surpasses existing judgment prediction benchmarks on both datasets, with GPT-4o mini achieving the highest performance (F1: 81.5 on PredEx, 80.3 on ILDC_expert), followed by Llama-3.1-8B. Human evaluation of the generated explanations across Clarity, Linking, and Usefulness metrics highlights GPT-4o mini's superior interpretability.

Title: Explaining AutoClustering: Uncovering Meta-Feature Contribution in AutoML for Clustering

• Authors: Matheus Camilo da Silva, Leonardo Arrighi, Ana Carolina Lorena, Sylvio Barbon Junior
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18348
• Pdf URL: https://arxiv.org/pdf/2602.18348
• Copy Paste: [[2602.18348]] Explaining AutoClustering: Uncovering Meta-Feature Contribution in AutoML for Clustering(https://arxiv.org/abs/2602.18348)
• Keywords: explainability
• Abstract: AutoClustering methods aim to automate unsupervised learning tasks, including algorithm selection (AS), hyperparameter optimization (HPO), and pipeline synthesis (PS), by often leveraging meta-learning over dataset meta-features. While these systems often achieve strong performance, their recommendations are often difficult to justify: the influence of dataset meta-features on algorithm and hyperparameter choices is typically not exposed, limiting reliability, bias diagnostics, and efficient meta-feature engineering. This limits reliability and diagnostic insight for further improvements. In this work, we investigate the explainability of the meta-models in AutoClustering. We first review 22 existing methods and organize their meta-features into a structured taxonomy. We then apply a global explainability technique (i.e., Decision Predicate Graphs) to assess feature importance within meta-models from selected frameworks. Finally, we use local explainability tools such as SHAP (SHapley Additive exPlanations) to analyse specific clustering decisions. Our findings highlight consistent patterns in meta-feature relevance, identify structural weaknesses in current meta-learning strategies that can distort recommendations, and provide actionable guidance for more interpretable Automated Machine Learning (AutoML) design. This study therefore offers a practical foundation for increasing decision transparency in unsupervised learning automation.

Title: Drawing the LINE: Cryptographic Analysis and Security Improvements for the LINE E2EE Protocol

• Authors: Benjamin Dowling, Prosanta Gope, Mehr U Nisa, Bhagya Wimalasiri
• Subjects: cs.CR
• Abstract URL: https://arxiv.org/abs/2602.18370
• Pdf URL: https://arxiv.org/pdf/2602.18370
• Copy Paste: [[2602.18370]] Drawing the LINE: Cryptographic Analysis and Security Improvements for the LINE E2EE Protocol(https://arxiv.org/abs/2602.18370)
• Keywords: security
• Abstract: LINE has emerged as one of the most popular communication platforms in many East Asian countries, including Thailand and Japan, with millions of active users. Therefore, it is essential to understand its security guarantees. In this work, we present the first provable security analysis of the LINE version two (LINEv2) messaging protocol, focusing on its cryptographic guarantees in a real-world setting. We capture the architecture and security of the LINE messaging protocol by modifying the Multi-Stage Key Exchange (MSKE) model, a framework for analysing cryptographic protocols under adversarial conditions. While LINEv2 achieves basic security properties such as key indistinguishability and message authentication, we highlight the lack of forward secrecy (FS) and post-compromise security (PCS). To address this, we introduce a stronger version of the LINE protocol, introducing FS and PCS to LINE, analysing and benchmarking our results.

Title: FedZMG: Efficient Client-Side Optimization in Federated Learning

• Authors: Fotios Zantalis, Evangelos Zervas, Grigorios Koulouras
• Subjects: cs.LG, cs.AI
• Abstract URL: https://arxiv.org/abs/2602.18384
• Pdf URL: https://arxiv.org/pdf/2602.18384
• Copy Paste: [[2602.18384]] FedZMG: Efficient Client-Side Optimization in Federated Learning(https://arxiv.org/abs/2602.18384)
• Keywords: privacy, federate
• Abstract: Federated Learning (FL) enables distributed model training on edge devices while preserving data privacy. However, clients tend to have non-Independent and Identically Distributed (non-IID) data, which often leads to client-drift, and therefore diminishing convergence speed and model performance. While adaptive optimizers have been proposed to mitigate these effects, they frequently introduce computational complexity or communication overhead unsuitable for resource-constrained IoT environments. This paper introduces Federated Zero Mean Gradients (FedZMG), a novel, parameter-free, client-side optimization algorithm designed to tackle client-drift by structurally regularizing the optimization space. Advancing the idea of Gradient Centralization, FedZMG projects local gradients onto a zero-mean hyperplane, effectively neutralizing the "intensity" or "bias" shifts inherent in heterogeneous data distributions without requiring additional communication or hyperparameter tuning. A theoretical analysis is provided, proving that FedZMG reduces the effective gradient variance and guarantees tighter convergence bounds compared to standard FedAvg. Extensive empirical evaluations on EMNIST, CIFAR100, and Shakespeare datasets demonstrate that FedZMG achieves better convergence speed and final validation accuracy compared to the baseline FedAvg and the adaptive optimizer FedAdam, particularly in highly non-IID settings.

Title: Self-Aware Object Detection via Degradation Manifolds

• Authors: Stefan Becker, Simon Weiss, Wolfgang Hübner, Michael Arens
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18394
• Pdf URL: https://arxiv.org/pdf/2602.18394
• Copy Paste: [[2602.18394]] Self-Aware Object Detection via Degradation Manifolds(https://arxiv.org/abs/2602.18394)
• Keywords: robust
• Abstract: Object detectors achieve strong performance under nominal imaging conditions but can fail silently when exposed to blur, noise, compression, adverse weather, or resolution changes. In safety-critical settings, it is therefore insufficient to produce predictions without assessing whether the input remains within the detector's nominal operating regime. We refer to this capability as self-aware object detection. We introduce a degradation-aware self-awareness framework based on degradation manifolds, which explicitly structure a detector's feature space according to image degradation rather than semantic content. Our method augments a standard detection backbone with a lightweight embedding head trained via multi-layer contrastive learning. Images sharing the same degradation composition are pulled together, while differing degradation configurations are pushed apart, yielding a geometrically organized representation that captures degradation type and severity without requiring degradation labels or explicit density modeling. To anchor the learned geometry, we estimate a pristine prototype from clean training embeddings, defining a nominal operating point in representation space. Self-awareness emerges as geometric deviation from this reference, providing an intrinsic, image-level signal of degradation-induced shift that is independent of detection confidence. Extensive experiments on synthetic corruption benchmarks, cross-dataset zero-shot transfer, and natural weather-induced distribution shifts demonstrate strong pristine-degraded separability, consistent behavior across multiple detector architectures, and robust generalization under semantic shift. These results suggest that degradation-aware representation geometry provides a practical and detector-agnostic foundation.

Title: PRISM-FCP: Byzantine-Resilient Federated Conformal Prediction via Partial Sharing

• Authors: Ehsan Lari, Reza Arablouei, Stefan Werner
• Subjects: cs.LG, eess.SP, math.PR, stat.AP, stat.ML
• Abstract URL: https://arxiv.org/abs/2602.18396
• Pdf URL: https://arxiv.org/pdf/2602.18396
• Copy Paste: [[2602.18396]] PRISM-FCP: Byzantine-Resilient Federated Conformal Prediction via Partial Sharing(https://arxiv.org/abs/2602.18396)
• Keywords: attack, robust, federate
• Abstract: We propose PRISM-FCP (Partial shaRing and robust calIbration with Statistical Margins for Federated Conformal Prediction), a Byzantine-resilient federated conformal prediction framework that utilizes partial model sharing to improve robustness against Byzantine attacks during both model training and conformal calibration. Existing approaches address adversarial behavior only in the calibration stage, leaving the learned model susceptible to poisoned updates. In contrast, PRISM-FCP mitigates attacks end-to-end. During training, clients partially share updates by transmitting only $M$ of $D$ parameters per round. This attenuates the expected energy of an adversary's perturbation in the aggregated update by a factor of $M/D$, yielding lower mean-square error (MSE) and tighter prediction intervals. During calibration, clients convert nonconformity scores into characterization vectors, compute distance-based maliciousness scores, and downweight or filter suspected Byzantine contributions before estimating the conformal quantile. Extensive experiments on both synthetic data and the UCI Superconductivity dataset demonstrate that PRISM-FCP maintains nominal coverage guarantees under Byzantine attacks while avoiding the interval inflation observed in standard FCP with reduced communication, providing a robust and communication-efficient approach to federated uncertainty quantification.

Title: Latent Equivariant Operators for Robust Object Recognition: Promise and Challenges

• Authors: Minh Dinh, Stéphane Deny
• Subjects: cs.CV, cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18406
• Pdf URL: https://arxiv.org/pdf/2602.18406
• Copy Paste: [[2602.18406]] Latent Equivariant Operators for Robust Object Recognition: Promise and Challenges(https://arxiv.org/abs/2602.18406)
• Keywords: robust
• Abstract: Despite the successes of deep learning in computer vision, difficulties persist in recognizing objects that have undergone group-symmetric transformations rarely seen during training-for example objects seen in unusual poses, scales, positions, or combinations thereof. Equivariant neural networks are a solution to the problem of generalizing across symmetric transformations, but require knowledge of transformations a priori. An alternative family of architectures proposes to earn equivariant operators in a latent space from examples of symmetric transformations. Here, using simple datasets of rotated and translated noisy MNIST, we illustrate how such architectures can successfully be harnessed for out-of-distribution classification, thus overcoming the limitations of both traditional and equivariant networks. While conceptually enticing, we discuss challenges ahead on the path of scaling these architectures to more complex datasets.

Title: Subgroups of $U(d)$ Induce Natural RNN and Transformer Architectures

• Authors: Joshua Nunley
• Subjects: cs.LG, cs.CL
• Abstract URL: https://arxiv.org/abs/2602.18417
• Pdf URL: https://arxiv.org/pdf/2602.18417
• Copy Paste: [[2602.18417]] Subgroups of $U(d)$ Induce Natural RNN and Transformer Architectures(https://arxiv.org/abs/2602.18417)
• Keywords: transformer
• Abstract: This paper presents a direct framework for sequence models with hidden states on closed subgroups of U(d). We use a minimal axiomatic setup and derive recurrent and transformer templates from a shared skeleton in which subgroup choice acts as a drop-in replacement for state space, tangent projection, and update map. We then specialize to O(d) and evaluate orthogonal-state RNN and transformer models on Tiny Shakespeare and Penn Treebank under parameter-matched settings. We also report a general linear-mixing extension in tangent space, which applies across subgroup choices and improves finite-budget performance in the current O(d) experiments.

Title: SPQ: An Ensemble Technique for Large Language Model Compression

• Authors: Jiamin Yao, Eren Gultepe
• Subjects: cs.CL
• Abstract URL: https://arxiv.org/abs/2602.18420
• Pdf URL: https://arxiv.org/pdf/2602.18420
• Copy Paste: [[2602.18420]] SPQ: An Ensemble Technique for Large Language Model Compression(https://arxiv.org/abs/2602.18420)
• Keywords: robust, large language model
• Abstract: This study presents an ensemble technique, SPQ (SVD-Pruning-Quantization), for large language model (LLM) compression that combines variance-retained singular value decomposition (SVD), activation-based pruning, and post-training linear quantization. Each component targets a different source of inefficiency: i) pruning removes redundant neurons in MLP layers, ii) SVD reduces attention projections into compact low-rank factors, iii) and 8-bit quantization uniformly compresses all linear layers. At matched compression ratios, SPQ outperforms individual methods (SVD-only, pruning-only, or quantization-only) in perplexity, demonstrating the benefit of combining complementary techniques. Applied to LLaMA-2-7B, SPQ achieves up to 75% memory reduction while maintaining or improving perplexity (e.g., WikiText-2 5.47 to 4.91) and preserving accuracy on downstream benchmarks such as C4, TruthfulQA, and GSM8K. Compared to strong baselines like GPTQ and SparseGPT, SPQ offers competitive perplexity and accuracy while using less memory (6.86 GB vs. 7.16 GB for GPTQ). Moreover, SPQ improves inference throughput over GPTQ, achieving up to a 1.9x speedup, which further enhances its practicality for real-world deployment. The effectiveness of SPQ's robust compression through layer-aware and complementary compression techniques may provide practical deployment of LLMs in memory-constrained environments. Code is available at: this https URL

Title: Generated Reality: Human-centric World Simulation using Interactive Video Generation with Hand and Camera Control

• Authors: Linxi Xie, Lisong C. Sun, Ashley Neall, Tong Wu, Shengqu Cai, Gordon Wetzstein
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18422
• Pdf URL: https://arxiv.org/pdf/2602.18422
• Copy Paste: [[2602.18422]] Generated Reality: Human-centric World Simulation using Interactive Video Generation with Hand and Camera Control(https://arxiv.org/abs/2602.18422)
• Keywords: diffusion, transformer, generative
• Abstract: Extended reality (XR) demands generative models that respond to users' tracked real-world motion, yet current video world models accept only coarse control signals such as text or keyboard input, limiting their utility for embodied interaction. We introduce a human-centric video world model that is conditioned on both tracked head pose and joint-level hand poses. For this purpose, we evaluate existing diffusion transformer conditioning strategies and propose an effective mechanism for 3D head and hand control, enabling dexterous hand--object interactions. We train a bidirectional video diffusion model teacher using this strategy and distill it into a causal, interactive system that generates egocentric virtual environments. We evaluate this generated reality system with human subjects and demonstrate improved task performance as well as a significantly higher level of perceived amount of control over the performed actions compared with relevant baselines.

Title: The Geometry of Noise: Why Diffusion Models Don't Need Noise Conditioning

• Authors: Mojtaba Sahraee-Ardakan, Mauricio Delbracio, Peyman Milanfar
• Subjects: cs.LG, cs.CV, eess.IV
• Abstract URL: https://arxiv.org/abs/2602.18428
• Pdf URL: https://arxiv.org/pdf/2602.18428
• Copy Paste: [[2602.18428]] The Geometry of Noise: Why Diffusion Models Don't Need Noise Conditioning(https://arxiv.org/abs/2602.18428)
• Keywords: diffusion, generative
• Abstract: Autonomous (noise-agnostic) generative models, such as Equilibrium Matching and blind diffusion, challenge the standard paradigm by learning a single, time-invariant vector field that operates without explicit noise-level conditioning. While recent work suggests that high-dimensional concentration allows these models to implicitly estimate noise levels from corrupted observations, a fundamental paradox remains: what is the underlying landscape being optimized when the noise level is treated as a random variable, and how can a bounded, noise-agnostic network remain stable near the data manifold where gradients typically diverge? We resolve this paradox by formalizing Marginal Energy, $E_{\text{marg}}(\mathbf{u}) = -\log p(\mathbf{u})$, where $p(\mathbf{u}) = \int p(\mathbf{u}|t)p(t)dt$ is the marginal density of the noisy data integrated over a prior distribution of unknown noise levels. We prove that generation using autonomous models is not merely blind denoising, but a specific form of Riemannian gradient flow on this Marginal Energy. Through a novel relative energy decomposition, we demonstrate that while the raw Marginal Energy landscape possesses a $1/t^p$ singularity normal to the data manifold, the learned time-invariant field implicitly incorporates a local conformal metric that perfectly counteracts the geometric singularity, converting an infinitely deep potential well into a stable attractor. We also establish the structural stability conditions for sampling with autonomous models. We identify a ``Jensen Gap'' in noise-prediction parameterizations that acts as a high-gain amplifier for estimation errors, explaining the catastrophic failure observed in deterministic blind models. Conversely, we prove that velocity-based parameterizations are inherently stable because they satisfy a bounded-gain condition that absorbs posterior uncertainty into a smooth geometric drift.

Title: VIRAASAT: Traversing Novel Paths for Indian Cultural Reasoning

• Authors: Harshul Raj Surana, Arijit Maji, Aryan Vats, Akash Ghosh, Sriparna Saha, Amit Sheth
• Subjects: cs.CL, cs.IR
• Abstract URL: https://arxiv.org/abs/2602.18429
• Pdf URL: https://arxiv.org/pdf/2602.18429
• Copy Paste: [[2602.18429]] VIRAASAT: Traversing Novel Paths for Indian Cultural Reasoning(https://arxiv.org/abs/2602.18429)
• Keywords: large language model
• Abstract: Large Language Models (LLMs) have made significant progress in reasoning tasks across various domains such as mathematics and coding. However, their performance deteriorates in tasks requiring rich socio-cultural knowledge and diverse local contexts, particularly those involving Indian Culture. Existing Cultural benchmarks are (i) Manually crafted, (ii) contain single-hop questions testing factual recall, and (iii) prohibitively costly to scale, leaving this deficiency largely unmeasured. To address this, we introduce VIRAASAT, a novel, semi-automated multi-hop approach for generating cultural specific multi-hop Question-Answering dataset for Indian culture. VIRAASAT leverages a Knowledge Graph comprising more than 700 expert-curated cultural artifacts, covering 13 key attributes of Indian culture (history, festivals, etc). VIRAASAT spans all 28 states and 8 Union Territories, yielding more than 3,200 multi-hop questions that necessitate chained cultural reasoning. We evaluate current State-of-the-Art (SOTA) LLMs on VIRAASAT and identify key limitations in reasoning wherein fine-tuning on Chain-of-Thought(CoT) traces fails to ground and synthesize low-probability facts. To bridge this gap, we propose a novel framework named Symbolic Chain-of-Manipulation (SCoM). Adapting the Chain-of-Manipulation paradigm, we train the model to simulate atomic Knowledge Graph manipulations internally. SCoM teaches the model to reliably traverse the topological structure of the graph. Experiments on Supervised Fine-Tuning (SFT) demonstrate that SCoM outperforms standard CoT baselines by up to 20%. We release the VIRAASAT dataset along with our findings, laying a strong foundation towards building Culturally Aware Reasoning Models.

Title: SARAH: Spatially Aware Real-time Agentic Humans

• Authors: Evonne Ng, Siwei Zhang, Zhang Chen, Michael Zollhoefer, Alexander Richard
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18432
• Pdf URL: https://arxiv.org/pdf/2602.18432
• Copy Paste: [[2602.18432]] SARAH: Spatially Aware Real-time Agentic Humans(https://arxiv.org/abs/2602.18432)
• Keywords: transformer
• Abstract: As embodied agents become central to VR, telepresence, and digital human applications, their motion must go beyond speech-aligned gestures: agents should turn toward users, respond to their movement, and maintain natural gaze. Current methods lack this spatial awareness. We close this gap with the first real-time, fully causal method for spatially-aware conversational motion, deployable on a streaming VR headset. Given a user's position and dyadic audio, our approach produces full-body motion that aligns gestures with speech while orienting the agent according to the user. Our architecture combines a causal transformer-based VAE with interleaved latent tokens for streaming inference and a flow matching model conditioned on user trajectory and audio. To support varying gaze preferences, we introduce a gaze scoring mechanism with classifier-free guidance to decouple learning from control: the model captures natural spatial alignment from data, while users can adjust eye contact intensity at inference time. On the Embody 3D dataset, our method achieves state-of-the-art motion quality at over 300 FPS -- 3x faster than non-causal baselines -- while capturing the subtle spatial dynamics of natural conversation. We validate our approach on a live VR system, bringing spatially-aware conversational agents to real-time deployment. Please see this https URL for details.

Title: Going Down Memory Lane: Scaling Tokens for Video Stream Understanding with Dynamic KV-Cache Memory

• Authors: Vatsal Agarwal, Saksham Suri, Matthew Gwilliam, Pulkit Kumar, Abhinav Shrivastava
• Subjects: cs.CV
• Abstract URL: https://arxiv.org/abs/2602.18434
• Pdf URL: https://arxiv.org/pdf/2602.18434
• Copy Paste: [[2602.18434]] Going Down Memory Lane: Scaling Tokens for Video Stream Understanding with Dynamic KV-Cache Memory(https://arxiv.org/abs/2602.18434)
• Keywords: robust
• Abstract: Streaming video understanding requires models to robustly encode, store, and retrieve information from a continuous video stream to support accurate video question answering (VQA). Existing state-of-the-art approaches rely on key-value caching to accumulate frame-level information over time, but use a limited number of tokens per frame, leading to the loss of fine-grained visual details. In this work, we propose scaling the token budget to enable more granular spatiotemporal understanding and reasoning. First, we find that current methods are ill-equipped to handle dense streams: their feature encoding causes query-frame similarity scores to increase over time, biasing retrieval toward later frames. To address this, we introduce an adaptive selection strategy that reduces token redundancy while preserving local spatiotemporal information. We further propose a training-free retrieval mixture-of-experts that leverages external models to better identify relevant frames. Our method, MemStream, achieves +8.0% on CG-Bench, +8.5% on LVBench, and +2.4% on VideoMME (Long) over ReKV with Qwen2.5-VL-7B.

Title: Assigning Confidence: K-partition Ensembles

• Authors: Aggelos Semoglou, John Pavlopoulos
• Subjects: cs.LG
• Abstract URL: https://arxiv.org/abs/2602.18435
• Pdf URL: https://arxiv.org/pdf/2602.18435
• Copy Paste: [[2602.18435]] Assigning Confidence: K-partition Ensembles(https://arxiv.org/abs/2602.18435)
• Keywords: robust
• Abstract: Clustering is widely used for unsupervised structure discovery, yet it offers limited insight into how reliable each individual assignment is. Diagnostics, such as convergence behavior or objective values, may reflect global quality, but they do not indicate whether particular instances are assigned confidently, especially for initialization-sensitive algorithms like k-means. This assignment-level instability can undermine both accuracy and robustness. Ensemble approaches improve global consistency by aggregating multiple runs, but they typically lack tools for quantifying pointwise confidence in a way that combines cross-run agreement with geometric support from the learned cluster structure. We introduce CAKE (Confidence in Assignments via K-partition Ensembles), a framework that evaluates each point using two complementary statistics computed over a clustering ensemble: assignment stability and consistency of local geometric fit. These are combined into a single, interpretable score in [0,1]. Our theoretical analysis shows that CAKE remains effective under noise and separates stable from unstable points. Experiments on synthetic and real-world datasets indicate that CAKE effectively highlights ambiguous points and stable core members, providing a confidence ranking that can guide filtering or prioritization to improve clustering quality.