# Research Papers — 2026-02 · bundle page 141 of 210 > Full-text markdown of papers #7001–7050 of 10493 published in 2026-02, newest first. AI-readable mirrors of the full 10493-paper 2026-02 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-02/page/141 [← Previous bundle](https://aipod.dev/papers/archive/2026-02/page/140.md) [Next bundle →](https://aipod.dev/papers/archive/2026-02/page/142.md) ### CTFExplorer: Evaluating LLM Offensive Agents Through Multi-Target Web CTF Benchmarking Authors: Nanda Rani, Kimberly Milner, Minghao Shao, Meet Udeshi, Haoran Xi, Venkata Sai Charan Putrevu, Saksham Aggarwal, Sandeep K. Shukla, Prashanth Krishnamurthy, Farshad Khorrami, Muhammad Shafique, Ramesh Karri Published: 2026-02-08 Categories: cs.CR, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2602.08023v3 Existing benchmarks for LLM-based offensive security agents use isolated, single-target setups with a known vulnerable service and fixed objective. They measure exploitation effectively, but miss how real Capture-the-Flag (CTF) participants triage unknown surfaces, prioritize targets, and allocate effort under uncertainty. Current evaluations therefore fail to assess strategic reasoning beyond exploitation alone. To address this, we introduce \textit{CTFExplorer}, a benchmark suite that shifts offensive security evaluation toward a multi-target setting, which tests how agents explore, prioritize, and chain attacks. CTFExplorer deploys 40 web-based vulnerable services within a single environment, where agents must autonomously discover, distinguish, and exploit targets without predefined guidance. We also present a reactive multi-agent setup as a reference agent framework and develop an agent-agnostic evaluation framework that records structured reasoning traces for fine-grained assessment. This enables behavioral evaluation beyond binary flag capture, such as how agents manage target selection, handle failed hypotheses, coordinate across multiple stages, and extract security intelligence. --- ### Structure-Aware Robust Counterfactual Explanations via Conditional Gaussian Network Classifiers Authors: Zhan-Yi Liao, Jaewon Yoo, Hao-Tsung Yang, Po-An Chen Published: 2026-02-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.08021v1 Counterfactual explanation (CE) is a core technique in explainable artificial intelligence (XAI), widely used to interpret model decisions and suggest actionable alternatives. This work presents a structure-aware and robustness-oriented counterfactual search method based on the conditional Gaussian network classifier (CGNC). The CGNC has a generative structure that encodes conditional dependencies and potential causal relations among features through a directed acyclic graph (DAG). This structure naturally embeds feature relationships into the search process, eliminating the need for additional constraints to ensure consistency with the model's structural assumptions. We adopt a convergence-guaranteed cutting-set procedure as an adversarial optimization framework, which iteratively approximates solutions that satisfy global robustness conditions. To address the nonconvex quadratic structure induced by feature dependencies, we apply piecewise McCormick relaxation to reformulate the problem as a mixed-integer linear program (MILP), ensuring global optimality. Experimental results show that our method achieves strong robustness, with direct global optimization of the original formulation providing especially stable and efficient results. The proposed framework is extensible to more complex constraint settings, laying the groundwork for future advances in counterfactual reasoning under nonconvex quadratic formulations. --- ### PhysDrape: Learning Explicit Forces and Collision Constraints for Physically Realistic Garment Draping Authors: Minghai Chen, Mingyuan Liu, Ning Ma, Jianqing Li, Yuxiang Huan Published: 2026-02-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.08020v2 Deep learning-based garment draping has emerged as a promising alternative to traditional Physics-Based Simulation (PBS), yet robust collision handling remains a critical bottleneck. Most existing methods enforce physical validity through soft penalties, creating an intrinsic trade-off between geometric feasibility and physical plausibility: penalizing collisions often distorts mesh structure, while preserving shape leads to interpenetration. To resolve this conflict, we present PhysDrape, a hybrid neural-physical solver for physically realistic garment draping driven by explicit forces and constraints. Unlike soft-constrained frameworks, PhysDrape integrates neural inference with explicit geometric solvers in a fully differentiable pipeline. Specifically, we propose a Physics-Informed Graph Neural Network conditioned on a physics-enriched graph -- encoding material parameters and body proximity -- to predict residual displacements. Crucially, we integrate a differentiable two-stage solver: first, a learnable Force Solver iteratively resolves unbalanced forces derived from the Saint Venant-Kirchhoff (StVK) model to ensure quasi-static equilibrium; second, a Differentiable Projection strictly enforces collision constraints against the body surface. This differentiable design guarantees physical validity through explicit constraints, while enabling end-to-end learning to optimize the network for physically consistent predictions. Extensive experiments demonstrate that PhysDrape achieves state-of-the-art performance, ensuring negligible interpenetration with significantly lower strain energy compared to existing baselines, achieving superior physical fidelity and robustness in real-time. --- ### The Rise of Sparse Mixture-of-Experts: A Survey from Algorithmic Foundations to Decentralized Architectures and Vertical Domain Applications Authors: Dong Pan, Bingtao Li, Yongsheng Zheng, Jiren Ma, Victor Fei Published: 2026-02-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.08019v1 The sparse Mixture of Experts(MoE) architecture has evolved as a powerful approach for scaling deep learning models to more parameters with comparable computation cost. As an important branch of large language model(LLM), MoE model only activate a subset of experts based on a routing network. This sparse conditional computation mechanism significantly improves computational efficiency, paving a promising path for greater scalability and cost-efficiency. It not only enhance downstream applications such as natural language processing, computer vision, and multimodal in various horizontal domains, but also exhibit broad applicability across vertical domains. Despite the growing popularity and application of MoE models across various domains, there lacks a systematic exploration of recent advancements of MoE in many important fields. Existing surveys on MoE suffer from limitations such as lack coverage or none extensively exploration of key areas. This survey seeks to fill these gaps. In this paper, Firstly, we examine the foundational principles of MoE, with an in-depth exploration of its core components-the routing network and expert network. Subsequently, we extend beyond the centralized paradigm to the decentralized paradigm, which unlocks the immense untapped potential of decentralized infrastructure, enables democratization of MoE development for broader communities, and delivers greater scalability and cost-efficiency. Furthermore we focus on exploring its vertical domain applications. Finally, we also identify key challenges and promising future research directions. To the best of our knowledge, this survey is currently the most comprehensive review in the field of MoE. We aim for this article to serve as a valuable resource for both researchers and practitioners, enabling them to navigate and stay up-to-date with the latest advancements. --- ### ICBAC: an Intelligent Contract-Based Access Control framework for supply chain management by integrating blockchain and federated learning Authors: Sadegh Sohani, Salar Ghazi, Farnaz Kamranfar, Sahar Pilehvar Moakhar, Mohammad Allahbakhsh, Haleh Amintoosi, Kaiwen Zhang Published: 2026-02-08 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2602.08014v1 This paper addresses the critical challenge of access control in modern supply chains, which operate across multiple independent and competing organizations. Existing access control is static and centralized, unable to adapt to insider threats or evolving contexts. Blockchain improves decentralization but lacks behavioral intelligence, while centralized machine learning for anomaly detection requires aggregating sensitive data, violating privacy. The proposed solution is ICBAC, an intelligent contract-based access control framework. It integrates permissioned blockchain (Hyperledger Fabric) with federated learning (FL). Built on Fabric, ICBAC uses a multi-channel architecture and three smart contracts for asset management, baseline access control, and dynamic revocation. To counter insider misuse, each channel deploys an AI agent that monitors activity and dynamically restricts access for anomalies. Federated learning allows these agents to collaboratively improve detection models without sharing raw data. For heterogeneous, competitive environments, ICBAC introduces a game-theoretic client selection mechanism using hedonic coalition formation. This enables supply chains to form stable, strategy-proof FL coalitions via preference-based selection without disclosing sensitive criteria. Extensive experiments on a Fabric testbed with a real-world dataset show ICBAC achieves blockchain performance comparable to static frameworks and provides effective anomaly detection under IID and non-IID data with zero raw-data sharing. ICBAC thus offers a practical, scalable solution for dynamic, privacy-preserving access control in decentralized supply chains. --- ### Small Agent Group is the Future of Digital Health Authors: Yuqiao Meng, Luoxi Tang, Dazheng Zhang, Rafael Brens, Elvys J. Romero, Nancy Guo, Safa Elkefi, Zhaohan Xi Published: 2026-02-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.08013v2 The rapid adoption of large language models (LLMs) in digital health has been driven by a "scaling-first" philosophy, i.e., the assumption that clinical intelligence increases with model size and data. However, real-world clinical needs include not only effectiveness, but also reliability and reasonable deployment cost. Since clinical decision-making is inherently collaborative, we challenge the monolithic scaling paradigm and ask whether a Small Agent Group (SAG) can support better clinical reasoning. SAG shifts from single-model intelligence to collective expertise by distributing reasoning, evidence-based analysis, and critical audit through a collaborative deliberation process. To assess the clinical utility of SAG, we conduct extensive evaluations using diverse clinical metrics spanning effectiveness, reliability, and deployment cost. Our results show that SAG achieves superior performance compared to a single giant model, both with and without additional optimization or retrieval-augmented generation. These findings suggest that the synergistic reasoning represented by SAG can substitute for model parameter growth in clinical settings. Overall, SAG offers a scalable solution to digital health that better balances effectiveness, reliability, and deployment efficiency. --- ### A Unified Density Operator View of Flow Control and Merging Authors: Riccardo De Santi, Malte Franke, Ya-Ping Hsieh, Andreas Krause Published: 2026-02-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.08012v1 Recent progress in large-scale flow and diffusion models raised two fundamental algorithmic challenges: (i) control-based reward adaptation of pre-trained flows, and (ii) integration of multiple models, i.e., flow merging. While current approaches address them separately, we introduce a unifying probability-space framework that subsumes both as limit cases, and enables reward-guided flow merging, allowing principled, task-aware combination of multiple pre-trained flows (e.g., merging priors while maximizing drug-discovery utilities). Our formulation renders possible to express a rich family of operators over generative models densities, including intersection (e.g., to enforce safety), union (e.g., to compose diverse models), interpolation (e.g., for discovery), their reward-guided counterparts, as well as complex logical expressions via generative circuits. Next, we introduce Reward-Guided Flow Merging (RFM), a mirror-descent scheme that reduces reward-guided flow merging to a sequence of standard fine-tuning problems. Then, we provide first-of-their-kind theoretical guarantees for reward-guided and pure flow merging via RFM. Ultimately, we showcase the capabilities of the proposed method on illustrative settings providing visually interpretable insights, and apply our method to high-dimensional de-novo molecular design and low-energy conformer generation. --- ### Towards Adaptive, Scalable, and Robust Coordination of LLM Agents: A Dynamic Ad-Hoc Networking Perspective Authors: Rui Li, Zeyu Zhang, Xiaohe Bo, Quanyu Dai, Chaozhuo Li, Feng Wen, Xu Chen Published: 2026-02-08 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2602.08009v1 Multi-agent architectures built on large language models (LLMs) have demonstrated the potential to realize swarm intelligence through well-crafted collaboration. However, the substantial burden of manual orchestration inherently raises an imperative to automate the design of agentic workflows. We frame such an agent coordination challenge as a classic problem in dynamic ad-hoc networking: How to establish adaptive and reliable communication among a scalable number of agentic hosts? In response to this unresolved dilemma, we introduce RAPS, a reputation-aware publish-subscribe paradigm for adaptive, scalable, and robust coordination of LLM agents. RAPS is grounded in the Distributed Publish-Subscribe Protocol, allowing LLM agents to exchange messages based on their declared intents rather than predefined topologies. Beyond this substrate, RAPS further incorporates two coherent overlays: (i) Reactive Subscription, enabling agents to dynamically refine their intents; and (ii) Bayesian Reputation, empowering each agent with a local watchdog to detect and isolate malicious peers. Extensive experiments over five benchmarks showcase that our design effectively reconciles adaptivity, scalability, and robustness in a unified multi-agent coordination framework. --- ### From $O(mn)$ to $O(r^2)$: Two-Sided Low-Rank Communication for Adam in Distributed Training with Memory Efficiency Authors: Sizhe Dang, Jiaqi Shao, Xiaodong Zheng, Guang Dai, Yan Song, Haishan Ye Published: 2026-02-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.08007v1 As foundation models continue to scale, pretraining increasingly relies on data-parallel distributed optimization, making bandwidth-limited gradient synchronization a key bottleneck. Orthogonally, projection-based low-rank optimizers were mainly designed for memory efficiency, but remain suboptimal for communication-limited training: one-sided synchronization still transmits an $O(rn)$ object for an $m\times n$ matrix gradient and refresh steps can dominate peak communicated bytes. We propose TSR, which brings two-sided low-rank communication to Adam-family updates (TSR-Adam) by synchronizing a compact core $U^\top G V\in\mathbb{R}^{r\times r}$, reducing the dominant per-step payload from $O(mn)$ to $O(r^2)$ while keeping moment states in low-dimensional cores. To further reduce the peak communication from subspace refresh, TSR-Adam adopts a randomized SVD-based refresh that avoids full-gradient synchronization. We additionally extend low-rank communication to embedding gradients with embedding-specific ranks and refresh schedules, yielding additional communication and memory savings over keeping embeddings dense. Across pretraining from 60M to 1B model scales, TSR-Adam reduces average communicated bytes per step by $13\times$, and on GLUE fine-tuning it reduces communication by $25\times$, while achieving comparable performance; we further provide a theoretical stationarity analysis for the proposed update. Code is available at https://github.com/DKmiyan/TSR-Adam. --- ### ForecastOcc: Vision-based Semantic Occupancy Forecasting Authors: Riya Mohan, Juana Valeria Hurtado, Rohit Mohan, Abhinav Valada Published: 2026-02-08 Categories: cs.CV, cs.AI, cs.LG, cs.RO arXiv: https://arxiv.org/abs/2602.08006v1 Autonomous driving requires forecasting both geometry and semantics over time to effectively reason about future environment states. Existing vision-based occupancy forecasting methods focus on motion-related categories such as static and dynamic objects, while semantic information remains largely absent. Recent semantic occupancy forecasting approaches address this gap but rely on past occupancy predictions obtained from separate networks. This makes current methods sensitive to error accumulation and prevents learning spatio-temporal features directly from images. In this work, we present ForecastOcc, the first framework for vision-based semantic occupancy forecasting that jointly predicts future occupancy states and semantic categories. Our framework yields semantic occupancy forecasts for multiple horizons directly from past camera images, without relying on externally estimated maps. We evaluate ForecastOcc in two complementary settings: multi-view forecasting on the Occ3D-nuScenes dataset and monocular forecasting on SemanticKITTI, where we establish the first benchmark for this task. We introduce the first baselines by adapting two 2D forecasting modules within our framework. Importantly, we propose a novel architecture that incorporates a temporal cross-attention forecasting module, a 2D-to-3D view transformer, a 3D encoder for occupancy prediction, and a semantic occupancy head for voxel-level forecasts across multiple horizons. Extensive experiments on both datasets show that ForecastOcc consistently outperforms baselines, yielding semantically rich, future-aware predictions that capture scene dynamics and semantics critical for autonomous driving. --- ### DeltaKV: Residual-Based KV Cache Compression via Long-Range Similarity Authors: Jitai Hao, Qiang Huang, Yaowei Wang, Min Zhang, Jun Yu Published: 2026-02-08 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.08005v1 The deployment of efficient long-context LLMs in applications like autonomous agents, long-chain reasoning, and creative writing is fundamentally bottlenecked by the linear growth of KV cache memory. Existing compression and eviction methods often struggle to balance accuracy, compression ratio, and hardware efficiency. We propose DeltaKV, a residual-based KV cache compression framework motivated by two empirical findings: long-range inter-token similarity and highly shared latent components in KV representations. Instead of discarding tokens, DeltaKV encodes semantic residuals relative to retrieved historical references, preserving fidelity while substantially reducing storage. To translate compression gains into real system speedups, we further introduce Sparse-vLLM, a high-performance inference engine with decoupled memory management and kernels optimized for sparse and irregular KV layouts. Experiments show that DeltaKV reduces KV cache memory to 29\% of the original while maintaining near-lossless accuracy on LongBench, SCBench, and AIME. When integrated with Sparse-vLLM, it achieves up to 2$\times$ throughput improvement over vLLM in long-context scenarios, demonstrating a practical path toward scalable long-context LLM deployment. Code, model checkpoints, and datasets are available at https://github.com/CURRENTF/Sparse-vLLM. --- ### Don't Always Pick the Highest-Performing Model: An Information Theoretic View of LLM Ensemble Selection Authors: Yigit Turkmen, Baturalp Buyukates, Melih Bastopcu Published: 2026-02-08 Categories: cs.LG, cs.AI, cs.DC, cs.IT, stat.ML arXiv: https://arxiv.org/abs/2602.08003v1 Large language models (LLMs) are often ensembled together to improve overall reliability and robustness, but in practice models are strongly correlated. This raises a fundamental question: which models should be selected when forming an LLM ensemble? We formulate budgeted ensemble selection as maximizing the mutual information between the true label and predictions of the selected models. Furthermore, to explain why performance can saturate even with many models, we model the correlated errors of the models using Gaussian-copula and show an information-theoretic error floor for the performance of the ensemble. Motivated by these, we propose a simple greedy mutual-information selection algorithm that estimates the required information terms directly from data and iteratively builds an ensemble under a query budget. We test our approach in two question answering datasets and one binary sentiment classification dataset: MEDMCQA, MMLU, and IMDB movie reviews. Across all datasets, we observe that our method consistently outperforms strong baselines under the same query budget. --- ### Regret Analysis of Unichain Average Reward Constrained MDPs with General Parameterization Authors: Anirudh Satheesh, Vaneet Aggarwal Published: 2026-02-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.08000v1 We study infinite-horizon average-reward constrained Markov decision processes (CMDPs) under the unichain assumption and general policy parameterizations. Existing regret analyses for constrained reinforcement learning largely rely on ergodicity or strong mixing-time assumptions, which fail to hold in the presence of transient states. We propose a primal--dual natural actor--critic algorithm that leverages multi-level Monte Carlo (MLMC) estimators and an explicit burn-in mechanism to handle unichain dynamics without requiring mixing-time oracles. Our analysis establishes finite-time regret and cumulative constraint violation bounds that scale as $\tilde{O}(\sqrt{T})$, up to approximation errors arising from policy and critic parameterization, thereby extending order-optimal guarantees to a significantly broader class of CMDPs. --- ### Tighter Information-Theoretic Generalization Bounds via a Novel Class of Change of Measure Inequalities Authors: Yanxiao Liu, Yijun Fan, Deniz Gündüz Published: 2026-02-08 Categories: cs.IT, cs.LG arXiv: https://arxiv.org/abs/2602.07999v4 Change of measure inequalities translate divergences between probability measures into explicit bounds on event probabilities, and play an important role in deriving probabilistic guarantees in learning theory, information theory, and statistics. We propose novel change of measure inequalities via a unified framework based on the data processing inequality, which is surprisingly elementary yet powerful enough to yield novel, tighter inequalities. We provide change of measure inequalities in terms of a broad family of information measures, including $f$-divergences (with Kullback-Leibler divergence and $χ^2$-divergence as special cases), Rényi divergence, and $α$-mutual information (with maximal leakage as a special case). We apply these results to generalization error analysis, PAC-Bayesian theory, differential privacy, and data memorization, obtaining stronger guarantees while recovering best-known results through simplified analyses. --- ### Fast Model Selection and Stable Optimization for Softmax-Gated Multinomial-Logistic Mixture of Experts Models Authors: TrungKhang Tran, TrungTin Nguyen, Md Abul Bashar, Nhat Ho, Richi Nayak, Christopher Drovandi Published: 2026-02-08 Categories: stat.ML, cs.LG, math.ST, stat.CO, stat.ME arXiv: https://arxiv.org/abs/2602.07997v1 Mixture-of-Experts (MoE) architectures combine specialized predictors through a learned gate and are effective across regression and classification, but for classification with softmax multinomial-logistic gating, rigorous guarantees for stable maximum-likelihood training and principled model selection remain limited. We address both issues in the full-data (batch) regime. First, we derive a batch minorization-maximization (MM) algorithm for softmax-gated multinomial-logistic MoE using an explicit quadratic minorizer, yielding coordinate-wise closed-form updates that guarantee monotone ascent of the objective and global convergence to a stationary point (in the standard MM sense), avoiding approximate M-steps common in EM-type implementations. Second, we prove finite-sample rates for conditional density estimation and parameter recovery, and we adapt dendrograms of mixing measures to the classification setting to obtain a sweep-free selector of the number of experts that achieves near-parametric optimal rates after merging redundant fitted atoms. Experiments on biological protein--protein interaction prediction validate the full pipeline, delivering improved accuracy and better-calibrated probabilities than strong statistical and machine-learning baselines. --- ### The Judge Who Never Admits: Hidden Shortcuts in LLM-based Evaluation Authors: Arash Marioriyad, Omid Ghahroodi, Ehsaneddin Asgari, Mohammad Hossein Rohban, Mahdieh Soleymani Baghshah Published: 2026-02-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.07996v1 Large language models (LLMs) are increasingly used as automatic judges to evaluate system outputs in tasks such as reasoning, question answering, and creative writing. A faithful judge should base its verdicts solely on content quality, remain invariant to irrelevant context, and transparently reflect the factors driving its decisions. We test this ideal via controlled cue perturbations-synthetic metadata labels injected into evaluation prompts-for six judge models: GPT-4o, Gemini-2.0-Flash, Gemma-3-27B, Qwen3-235B, Claude-3-Haiku, and Llama3-70B. Experiments span two complementary datasets with distinct evaluation regimes: ELI5 (factual QA) and LitBench (open-ended creative writing). We study six cue families: source, temporal, age, gender, ethnicity, and educational status. Beyond measuring verdict shift rates (VSR), we introduce cue acknowledgment rate (CAR) to quantify whether judges explicitly reference the injected cues in their natural-language rationales. Across cues with strong behavioral effects-e.g., provenance hierarchies (Expert > Human > LLM > Unknown), recency preferences (New > Old), and educational-status favoritism-CAR is typically at or near zero, indicating that shortcut reliance is largely unreported even when it drives decisions. Crucially, CAR is also dataset-dependent: explicit cue recognition is more likely to surface in the factual ELI5 setting for some models and cues, but often collapses in the open-ended LitBench regime, where large verdict shifts can persist despite zero acknowledgment. The combination of substantial verdict sensitivity and limited cue acknowledgment reveals an explanation gap in LLM-as-judge pipelines, raising concerns about reliability of model-based evaluation in both research and deployment. --- ### MCIE: Multimodal LLM-Driven Complex Instruction Image Editing with Spatial Guidance Authors: Xuehai Bai, Xiaoling Gu, Akide Liu, Hangjie Yuan, YiFan Zhang, Jack Ma Published: 2026-02-08 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.07993v1 Recent advances in instruction-based image editing have shown remarkable progress. However, existing methods remain limited to relatively simple editing operations, hindering real-world applications that require complex and compositional instructions. In this work, we address these limitations from the perspectives of architectural design, data, and evaluation protocols. Specifically, we identify two key challenges in current models: insufficient instruction compliance and background inconsistency. To this end, we propose MCIE-E1, a Multimodal Large Language Model-Driven Complex Instruction Image Editing method that integrates two key modules: a spatial-aware cross-attention module and a background-consistent cross-attention module. The former enhances instruction-following capability by explicitly aligning semantic instructions with spatial regions through spatial guidance during the denoising process, while the latter preserves features in unedited regions to maintain background consistency. To enable effective training, we construct a dedicated data pipeline to mitigate the scarcity of complex instruction-based image editing datasets, combining fine-grained automatic filtering via a powerful MLLM with rigorous human validation. Finally, to comprehensively evaluate complex instruction-based image editing, we introduce CIE-Bench, a new benchmark with two new evaluation metrics. Experimental results on CIE-Bench demonstrate that MCIE-E1 consistently outperforms previous state-of-the-art methods in both quantitative and qualitative assessments, achieving a 23.96% improvement in instruction compliance. --- ### When Is Compositional Reasoning Learnable from Verifiable Rewards? Authors: Daniel Barzilai, Yotam Wolf, Ronen Basri Published: 2026-02-08 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2602.07992v1 The emergence of compositional reasoning in large language models through reinforcement learning with verifiable rewards (RLVR) has been a key driver of recent empirical successes. Despite this progress, it remains unclear which compositional problems are learnable in this setting using outcome-level feedback alone. In this work, we theoretically study the learnability of compositional problems in autoregressive models under RLVR training. We identify a quantity that we call the task-advantage ratio, a joint property of the compositional problem and the base model, that characterizes which tasks and compositions are learnable from outcome-level feedback. On the positive side, using this characterization, we show that compositional problems where correct intermediate steps provide a clear advantage are efficiently learnable with RLVR. We also analyze how such an advantage naturally arises in different problems. On the negative side, when the structural advantage is not present, RLVR may converge to suboptimal compositions. We prove that, in some cases, the quality of the base model determines if such an advantage exists and whether RLVR will converge to a suboptimal solution. We hope our analysis can provide a principled theoretical understanding of when and why RLVR succeeds and when it does not. --- ### Learning to Alleviate Familiarity Bias in Video Recommendation Authors: Zheng Ren, Yi Wu, Jianan Lu, Acar Ary, Yiqu Liu, Li Wei, Lukasz Heldt Published: 2026-02-08 Categories: cs.IR, cs.LG arXiv: https://arxiv.org/abs/2602.07987v1 Modern video recommendation systems aim to optimize user engagement and platform objectives, yet often face structural exposure imbalances caused by behavioral biases. In this work, we focus on the post-ranking stage and present LAFB (Learning to Alleviate Familiarity Bias), a lightweight and model-agnostic framework designed to mitigate familiarity bias in recommendation outputs. LAFB models user-content familiarity using discrete and continuous interaction features, and estimates personalized debiasing factors to adjust user rating prediction scores, thereby reducing the dominance of familiar content in the final ranking. We conduct large-scale offline evaluations and online A/B testing in a real-world recommendation system, under a unified serving stack that also compares LAFB with deployable popularity-oriented remedies. Results show that LAFB increases novel watch-time share and improves exposure for emerging creators and overall content diversity, while maintaining stable overall watch time and short-term satisfaction. LAFB has already been launched in the post-ranking stage of YouTube's recommendation system, demonstrating its effectiveness in real-world applications. --- ### Deepfake Synthesis vs. Detection: An Uneven Contest Authors: Md. Tarek Hasan, Sanjay Saha, Shaojing Fan, Swakkhar Shatabda, Terence Sim Published: 2026-02-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.07986v1 The rapid advancement of deepfake technology has significantly elevated the realism and accessibility of synthetic media. Emerging techniques, such as diffusion-based models and Neural Radiance Fields (NeRF), alongside enhancements in traditional Generative Adversarial Networks (GANs), have contributed to the sophisticated generation of deepfake videos. Concurrently, deepfake detection methods have seen notable progress, driven by innovations in Transformer architectures, contrastive learning, and other machine learning approaches. In this study, we conduct a comprehensive empirical analysis of state-of-the-art deepfake detection techniques, including human evaluation experiments against cutting-edge synthesis methods. Our findings highlight a concerning trend: many state-of-the-art detection models exhibit markedly poor performance when challenged with deepfakes produced by modern synthesis techniques, including poor performance by human participants against the best quality deepfakes. Through extensive experimentation, we provide evidence that underscores the urgent need for continued refinement of detection models to keep pace with the evolving capabilities of deepfake generation technologies. This research emphasizes the critical gap between current detection methodologies and the sophistication of new generation techniques, calling for intensified efforts in this crucial area of study. --- ### Accelerating Social Science Research via Agentic Hypothesization and Experimentation Authors: Jishu Sen Gupta, Harini SI, Somesh Kumar Singh, Syed Mohamad Tawseeq, Yaman Kumar Singla, David Doermann, Rajiv Ratn Shah, Balaji Krishnamurthy Published: 2026-02-08 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2602.07983v1 Data-driven social science research is inherently slow, relying on iterative cycles of observation, hypothesis generation, and experimental validation. While recent data-driven methods promise to accelerate parts of this process, they largely fail to support end-to-end scientific discovery. To address this gap, we introduce EXPERIGEN, an agentic framework that operationalizes end-to-end discovery through a Bayesian optimization inspired two-phase search, in which a Generator proposes candidate hypotheses and an Experimenter evaluates them empirically. Across multiple domains, EXPERIGEN consistently discovers 2-4x more statistically significant hypotheses that are 7-17 percent more predictive than prior approaches, and naturally extends to complex data regimes including multimodal and relational datasets. Beyond statistical performance, hypotheses must be novel, empirically grounded, and actionable to drive real scientific progress. To evaluate these qualities, we conduct an expert review of machine-generated hypotheses, collecting feedback from senior faculty. Among 25 reviewed hypotheses, 88 percent were rated moderately or strongly novel, 70 percent were deemed impactful and worth pursuing, and most demonstrated rigor comparable to senior graduate-level research. Finally, recognizing that ultimate validation requires real-world evidence, we conduct the first A/B test of LLM-generated hypotheses, observing statistically significant results with p less than 1e-6 and a large effect size of 344 percent. --- ### Continuity-driven Synergistic Diffusion with Neural Priors for Ultra-Sparse-View CBCT Reconstruction Authors: Junlin Wang, Jiancheng Fang, Peng Peng, Shaoyu Wang, Qiegen Liu Published: 2026-02-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.07980v1 The clinical application of cone-beam computed tomography (CBCT) is constrained by the inherent trade-off between radiation exposure and image quality. Ultra-sparse angular sampling, employed to reduce dose, introduces severe undersampling artifacts and inter-slice inconsistencies, compromising diagnostic reliability. Existing reconstruction methods often struggle to balance angular continuity with spatial detail fidelity. To address these challenges, we propose a Continuity-driven Synergistic Diffusion with Neural priors (CSDN) for ultra-sparse-view CBCT reconstruction. Neural priors are introduced as a structural foundation to encode a continuous threedimensional attenuation representation, enabling the synthesis of physically consistent dense projections from ultra-sparse measurements. Building upon this neural-prior-based initialization, a synergistic diffusion strategy is developed, consisting of two collaborative refinement paths: a Sinogram Refinement Diffusion (Sino-RD) process that restores angular continuity and a Digital Radiography Refinement Diffusion (DR-RD) process that enforces inter-slice consistency from the projection image perspective. The outputs of the two diffusion paths are adaptively fused by the Dual-Projection Reconstruction Fusion (DPRF) module to achieve coherent volumetric reconstruction. Extensive experiments demonstrate that the proposed CSDN effectively suppresses artifacts and recovers fine textures under ultra-sparse-view conditions, outperforming existing state-of-the-art techniques. --- ### FSP-Diff: Full-Spectrum Prior-Enhanced DualDomain Latent Diffusion for Ultra-Low-Dose Spectral CT Reconstruction Authors: Peng Peng, Xinrui Zhang, Junlin Wang, Lei Li, Shaoyu Wang, Qiegen Liu Published: 2026-02-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.07979v1 Spectral computed tomography (CT) with photon-counting detectors holds immense potential for material discrimination and tissue characterization. However, under ultra-low-dose conditions, the sharply degraded signal-to-noise ratio (SNR) in energy-specific projections poses a significant challenge, leading to severe artifacts and loss of structural details in reconstructed images. To address this, we propose FSP-Diff, a full-spectrum prior-enhanced dual-domain latent diffusion framework for ultra-low-dose spectral CT reconstruction. Our framework integrates three core strategies: 1) Complementary Feature Construction: We integrate direct image reconstructions with projection-domain denoised results. While the former preserves latent textural nuances amidst heavy noise, the latter provides a stable structural scaffold to balance detail fidelity and noise suppression. 2) Full-Spectrum Prior Integration: By fusing multi-energy projections into a high-SNR full-spectrum image, we establish a unified structural reference that guides the reconstruction across all energy bins. 3) Efficient Latent Diffusion Synthesis: To alleviate the high computational burden of high-dimensional spectral data, multi-path features are embedded into a compact latent space. This allows the diffusion process to facilitate interactive feature fusion in a lower-dimensional manifold, achieving accelerated reconstruction while maintaining fine-grained detail restoration. Extensive experiments on simulated and real-world datasets demonstrate that FSP-Diff significantly outperforms state-of-the-art methods in both image quality and computational efficiency, underscoring its potential for clinically viable ultra-low-dose spectral CT imaging. --- ### Cross-Linguistic Persona-Driven Data Synthesis for Robust Multimodal Cognitive Decline Detection Authors: Rui Feng, Zhiyao Luo, Liuyu Wu, Wei Wang, Yuting Song, Yong Liu, Kok Pin Ng, Jianqing Li, Xingyao Wang Published: 2026-02-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.07978v1 Speech-based digital biomarkers represent a scalable, non-invasive frontier for the early identification of Mild Cognitive Impairment (MCI). However, the development of robust diagnostic models remains impeded by acute clinical data scarcity and a lack of interpretable reasoning. Current solutions frequently struggle with cross-lingual generalization and fail to provide the transparent rationales essential for clinical trust. To address these barriers, we introduce SynCog, a novel framework integrating controllable zero-shot multimodal data synthesis with Chain-of-Thought (CoT) deduction fine-tuning. Specifically, SynCog simulates diverse virtual subjects with varying cognitive profiles to effectively alleviate clinical data scarcity. This generative paradigm enables the rapid, zero-shot expansion of clinical corpora across diverse languages, effectively bypassing data bottlenecks in low-resource settings and bolstering the diagnostic performance of Multimodal Large Language Models (MLLMs). Leveraging this synthesized dataset, we fine-tune a foundational multimodal backbone using a CoT deduction strategy, empowering the model to explicitly articulate diagnostic thought processes rather than relying on black-box predictions. Extensive experiments on the ADReSS and ADReSSo benchmarks demonstrate that augmenting limited clinical data with synthetic phenotypes yields competitive diagnostic performance, achieving Macro-F1 scores of 80.67% and 78.46%, respectively, outperforming current baseline models. Furthermore, evaluation on an independent real-world Mandarin cohort (CIR-E) demonstrates robust cross-linguistic generalization, attaining a Macro-F1 of 48.71%. These findings constitute a critical step toward providing clinically trustworthy and linguistically inclusive cognitive assessment tools for global healthcare. --- ### Structural Learning Theory: A Metric-Topology Factorization Approach Authors: Xin Li Published: 2026-02-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.07974v2 Learning in structured, multi-context, or non-stationary environments involves two orthogonal difficulties. The first is \emph{metric}: once the correct context is known, how hard is prediction within it? This is the domain of Statistical Learning Theory (SLT). The second is \emph{structural}: how many local contexts are required, and how can they be discovered from data? This paper develops \emph{Structural Learning Theory} (StrLT) for the structural axis. We introduce \emph{width}, the minimum number of jointly contractive and low-risk cells needed to cover a learning problem. Width is incomparable with VC dimension: either can diverge while the other remains bounded. We show that width induces a \emph{phase transition}: if the allocated number of cells \(K<w\), learning suffers an irreducible structural error floor; if \(K\ge w\), the problem reduces to ordinary within-cell statistical learning. To estimate width, we introduce the \emph{contractive-similarity} (CS) operator, a task-adaptive graph kernel combining geometric locality with predictive compatibility. Its CS Laplacian exposes contractive basins through spectral separation. We further develop the \emph{metric slingshot}, which reuses low-dimensional latent contraction maps to reduce funnel-learning cost. Together, width, CS estimation, and the slingshot decompose learning into trap discovery and funnel generalization, with deep implications for continual and lifelong learning in an open-ended environment. --- ### On Improving Neurosymbolic Learning by Exploiting the Representation Space Authors: Aaditya Naik, Efthymia Tsamoura, Shibo Jin, Mayur Naik, Dan Roth Published: 2026-02-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.07973v1 We study the problem of learning neural classifiers in a neurosymbolic setting where the hidden gold labels of input instances must satisfy a logical formula. Learning in this setting proceeds by first computing (a subset of) the possible combinations of labels that satisfy the formula and then computing a loss using those combinations and the classifiers' scores. One challenge is that the space of label combinations can grow exponentially, making learning difficult. We propose a technique that prunes this space by exploiting the intuition that instances with similar latent representations are likely to share the same label. While this intuition has been widely used in weakly supervised learning, its application in our setting is challenging due to label dependencies imposed by logical constraints. We formulate the pruning process as an integer linear program that discards inconsistent label combinations while respecting logical structure. Our approach, CLIPPER, is orthogonal to existing training algorithms and can be seamlessly integrated with them. Across 16 benchmarks over complex neurosymbolic tasks, we demonstrate that CLIPPER boosts the performance of state-of-the-art neurosymbolic engines like Scallop, Dolphin, and ISED by up to 48%, 53%, and 8%, leading to state-of-the-art accuracies. --- ### Learning-guided Kansa collocation for forward and inverse PDEs beyond linearity Authors: Zheyuan Hu, Weitao Chen, Cengiz Öztireli, Chenliang Zhou, Fangcheng Zhong Published: 2026-02-08 Categories: cs.CE, cs.AI, cs.LG, math.NA arXiv: https://arxiv.org/abs/2602.07970v3 Partial Differential Equations are precise in modelling the physical, biological and graphical phenomena. However, the numerical methods suffer from the curse of dimensionality, high computation costs and domain-specific discretization. We aim to explore pros and cons of different PDE solvers, and apply them to specific scientific simulation problems, including forwarding solution, inverse problems and equations discovery. In particular, we extend the recent CNF (NeurIPS 2023) framework solver to coupled and non-linear settings, together with down-stream applications. The outcomes include implementation of selected methods, self-tuning techniques, evaluation on benchmark problems and a comprehensive survey of neural PDE solvers and scientific simulation applications. --- ### EasyTune: Efficient Step-Aware Fine-Tuning for Diffusion-Based Motion Generation Authors: Xiaofeng Tan, Wanjiang Weng, Haodong Lei, Hongsong Wang Published: 2026-02-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.07967v1 In recent years, motion generative models have undergone significant advancement, yet pose challenges in aligning with downstream objectives. Recent studies have shown that using differentiable rewards to directly align the preference of diffusion models yields promising results. However, these methods suffer from (1) inefficient and coarse-grained optimization with (2) high memory consumption. In this work, we first theoretically and empirically identify the key reason of these limitations: the recursive dependence between different steps in the denoising trajectory. Inspired by this insight, we propose EasyTune, which fine-tunes diffusion at each denoising step rather than over the entire trajectory. This decouples the recursive dependence, allowing us to perform (1) a dense and fine-grained, and (2) memory-efficient optimization. Furthermore, the scarcity of preference motion pairs restricts the availability of motion reward model training. To this end, we further introduce a Self-refinement Preference Learning (SPL) mechanism that dynamically identifies preference pairs and conducts preference learning. Extensive experiments demonstrate that EasyTune outperforms DRaFT-50 by 8.2% in alignment (MM-Dist) improvement while requiring only 31.16% of its additional memory overhead and achieving a 7.3x training speedup. The project page is available at this link {https://xiaofeng-tan.github.io/projects/EasyTune/index.html}. --- ### An Explainable Multi-Task Similarity Measure: Integrating Accumulated Local Effects and Weighted Fréchet Distance Authors: Pablo Hidalgo, Daniel Rodriguez Published: 2026-02-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.07966v1 In many machine learning contexts, tasks are often treated as interconnected components with the goal of leveraging knowledge transfer between them, which is the central aim of Multi-Task Learning (MTL). Consequently, this multi-task scenario requires addressing critical questions: which tasks are similar, and how and why do they exhibit similarity? In this work, we propose a multi-task similarity measure based on Explainable Artificial Intelligence (XAI) techniques, specifically Accumulated Local Effects (ALE) curves. ALE curves are compared using the Fréchet distance, weighted by the data distribution, and the resulting similarity measure incorporates the importance of each feature. The measure is applicable in both single-task learning scenarios, where each task is trained separately, and multi-task learning scenarios, where all tasks are learned simultaneously. The measure is model-agnostic, allowing the use of different machine learning models across tasks. A scaling factor is introduced to account for differences in predictive performance across tasks, and several recommendations are provided for applying the measure in complex scenarios. We validate this measure using four datasets, one synthetic dataset and three real-world datasets. The real-world datasets include a well-known Parkinson's dataset and a bike-sharing usage dataset -- both structured in tabular format -- as well as the CelebA dataset, which is used to evaluate the application of concept bottleneck encoders in a multitask learning setting. The results demonstrate that the measure aligns with intuitive expectations of task similarity across both tabular and non-tabular data, making it a valuable tool for exploring relationships between tasks and supporting informed decision-making. --- ### Lost in Translation? A Comparative Study on the Cross-Lingual Transfer of Composite Harms Authors: Vaibhav Shukla, Hardik Sharma, Adith N Reganti, Soham Wasmatkar, Bagesh Kumar, Vrijendra Singh Published: 2026-02-08 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.07963v1 Most safety evaluations of large language models (LLMs) remain anchored in English. Translation is often used as a shortcut to probe multilingual behavior, but it rarely captures the full picture, especially when harmful intent or structure morphs across languages. Some types of harm survive translation almost intact, while others distort or disappear. To study this effect, we introduce CompositeHarm, a translation-based benchmark designed to examine how safety alignment holds up as both syntax and semantics shift. It combines two complementary English datasets, AttaQ, which targets structured adversarial attacks, and MMSafetyBench, which covers contextual, real-world harms, and extends them into six languages: English, Hindi, Assamese, Marathi, Kannada, and Gujarati. Using three large models, we find that attack success rates rise sharply in Indic languages, especially under adversarial syntax, while contextual harms transfer more moderately. To ensure scalability and energy efficiency, our study adopts lightweight inference strategies inspired by edge-AI design principles, reducing redundant evaluation passes while preserving cross-lingual fidelity. This design makes large-scale multilingual safety testing both computationally feasible and environmentally conscious. Overall, our results show that translated benchmarks are a necessary first step, but not a sufficient one, toward building grounded, resource-aware, language-adaptive safety systems. --- ### LOCA-bench: Benchmarking Language Agents Under Controllable and Extreme Context Growth Authors: Weihao Zeng, Yuzhen Huang, Junxian He Published: 2026-02-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.07962v1 Large language models (LLMs) are increasingly capable of carrying out long-running, real-world tasks. However, as the amount of context grows, their reliability often deteriorates, a phenomenon known as "context rot". Existing long-context benchmarks primarily focus on single-step settings that evaluate a model's ability to retrieve information from a long snippet. In realistic scenarios, however, LLMs often need to act as agents that explore environments, follow instructions and plans, extract useful information, and predict correct actions under a dynamically growing context. To assess language agents in such settings, we introduce LOCA-bench (a benchmark for LOng-Context Agents). Given a task prompt, LOCA-bench leverages automated and scalable control of environment states to regulate the agent's context length. This design enables LOCA-bench to extend the context length potentially to infinity in a controlled way while keeping the underlying task semantics fixed. LOCA-bench evaluates language agents as a combination of models and scaffolds, including various context management strategies. While agent performance generally degrades as the environment states grow more complex, advanced context management techniques can substantially improve the overall success rate. We open-source LOCA-bench to provide a platform for evaluating models and scaffolds in long-context, agentic scenarios: https://github.com/hkust-nlp/LOCA-bench --- ### D-ORCA: Dialogue-Centric Optimization for Robust Audio-Visual Captioning Authors: Changli Tang, Tianyi Wang, Fengyun Rao, Jing Lyu, Chao Zhang Published: 2026-02-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.07960v1 Spoken dialogue is a primary source of information in videos; therefore, accurately identifying who spoke what and when is essential for deep video understanding. We introduce D-ORCA, a \textbf{d}ialogue-centric \textbf{o}mni-modal large language model optimized for \textbf{r}obust audio-visual \textbf{ca}ptioning. We further curate DVD, a large-scale, high-quality bilingual dataset comprising nearly 40,000 multi-party dialogue videos for training and 2000 videos for evaluation in English and Mandarin, addressing a critical gap in the open-source ecosystem. To ensure fine-grained captioning accuracy, we adopt group relative policy optimization with three novel reward functions that assess speaker attribution accuracy, global speech content accuracy, and sentence-level temporal boundary alignment. These rewards are derived from evaluation metrics widely used in speech processing and, to our knowledge, are applied for the first time as reinforcement learning objectives for audio-visual captioning. Extensive experiments demonstrate that D-ORCA substantially outperforms existing open-source models in speaker identification, speech recognition, and temporal grounding. Notably, despite having only 8 billion parameters, D-ORCA achieves performance competitive with Qwen3-Omni across several general-purpose audio-visual understanding benchmarks. Demos are available at \href{https://d-orca-llm.github.io/}{https://d-orca-llm.github.io/}. Our code, data, and checkpoints will be available at \href{https://github.com/WeChatCV/D-ORCA/}{https://github.com/WeChatCV/D-ORCA/}. --- ### Accuracy-Delay Trade-Off in LLM Offloading via Token-Level Uncertainty Authors: Yumin Kim, Hyeonsu Lyu, Minjae Lee, Hyun Jong Yang Published: 2026-02-08 Categories: eess.SY, cs.AI, cs.DC arXiv: https://arxiv.org/abs/2602.07958v1 Large language models (LLMs) offer significant potential for intelligent mobile services but are computationally intensive for resource-constrained devices. Mobile edge computing (MEC) allows such devices to offload inference tasks to edge servers (ESs), yet introduces latency due to communication and serverside queuing, especially in multi-user environments. In this work, we propose an uncertainty-aware offloading framework that dynamically decides whether to perform inference locally or offload it to the ES, based on token-level uncertainty and resource constraints. We define a margin-based token-level uncertainty metric and demonstrate its correlation with model accuracy. Leveraging this metric, we design a greedy offloading algorithm (GOA) that minimizes delay while maintaining accuracy by prioritizing offloading for highuncertainty queries. Our experiments show that GOA consistently achieves a favorable trade-off, outperforming baseline strategies in both accuracy and latency across varying user densities, and operates with practical computation time. These results establish GOA as a scalable and effective solution for LLM inference in MEC environments. --- ### One-Shot Crowd Counting With Density Guidance For Scene Adaptation Authors: Jiwei Chen, Qi Wang, Junyu Gao, Jing Zhang, Dingyi Li, Jing-Jia Luo Published: 2026-02-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.07955v2 Crowd scenes captured by cameras at different locations vary greatly, and existing crowd models have limited generalization for unseen surveillance scenes. To improve the generalization of the model, we regard different surveillance scenes as different category scenes, and introduce few-shot learning to make the model adapt to the unseen surveillance scene that belongs to the given exemplar category scene. To this end, we propose to leverage local and global density characteristics to guide the model of crowd counting for unseen surveillance scenes. Specifically, to enable the model to adapt to the varying density variations in the target scene, we propose the multiple local density learner to learn multi prototypes which represent different density distributions in the support scene. Subsequently, these multiple local density similarity matrixes are encoded. And they are utilized to guide the model in a local way. To further adapt to the global density in the target scene, the global density features are extracted from the support image, then it is used to guide the model in a global way. Experiments on three surveillance datasets shows that proposed method can adapt to the unseen surveillance scene and outperform recent state-of-the-art methods in the few-shot crowd counting. --- ### Bielik Guard: Efficient Polish Language Safety Classifiers for LLM Content Moderation Authors: Krzysztof Wróbel, Jan Maria Kowalski, Jerzy Surma, Igor Ciuciura, Maciej Szymański Published: 2026-02-08 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.07954v4 As Large Language Models (LLMs) become increasingly deployed in Polish language applications, the need for efficient and accurate content safety classifiers has become paramount. We present Bielik Guard, a family of compact Polish language safety classifiers comprising two model variants: a 0.1B parameter model based on MMLW-RoBERTa-base and a 0.5B parameter model based on PKOBP/polish-roberta-8k. Fine-tuned on a community-annotated dataset of 6,885 Polish texts, these models classify content across five safety categories: Hate/Aggression, Vulgarities, Sexual Content, Crime, and Self-Harm. Our evaluation demonstrates that both models achieve strong performance on multiple benchmarks. The 0.5B variant offers the best overall discrimination capability with F1 scores of 0.791 (micro) and 0.785 (macro) on the test set, while the 0.1B variant demonstrates exceptional efficiency. Notably, Bielik Guard 0.1B v1.1 achieves superior precision (77.65%) and very low false positive rate (0.63%) on real user prompts, outperforming HerBERT-PL-Guard (31.55% precision, 4.70% FPR) despite identical model size. The models are publicly available and designed to provide appropriate responses rather than simple content blocking, particularly for sensitive categories like self-harm. --- ### A Thermodynamic Theory of Learning Part II: History-Dependent Reachability and Continual Learning Authors: Daisuke Okanohara Published: 2026-02-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.07950v3 We formulate plasticity as target-dependent, finite-horizon reachability under history-dependent dynamics, using standard minimum-energy control theory. An extended state includes parameters and internal variables that affect future updates. Around a reference trajectory, the input-to-endpoint response and its controllability Gramian determine the minimum input energy for each reachable displacement. Minimizing this energy over the joint task target defines a task-conditioned adaptation cost, exact for the arbitrary-input linear model. A solvable learning-retention example shows that this cost can increase without any rank loss, and distinguishes small directional gain from an inaccessible direction. Wasserstein transport supplies a complementary global lower bound, while learning-map Jacobians describe deformation of initial perturbations rather than response to future inputs. The resulting learning-retention frontier depends on the current extended state, admissible updates, target, horizon, and cost metric; nonlinear and constrained-update applications require additional control of these modeling choices. --- ### A Thermodynamic Theory of Learning Part II: Critical Period Closure and Continual Learning Failure Authors: Daisuke Okanohara Published: 2026-02-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.07950v2 Learning performed over finite time is inherently irreversible. In Part~I of this series, we modeled learning as a transport process in the space of parameter distributions and derived the Epistemic Speed Limit (ESL), which lower-bounds entropy production under finite-time dynamics. In this work (Part~II), we show that irreversibility imposes a geometric restriction on future adaptability through the compositional structure of learning dynamics. Successive learning phases compose multiplicatively as transport maps, and their Jacobians form a semigroup whose rank and singular values are submultiplicative. As a result, dynamically usable degrees of reconfiguration can only decrease under composition. We formalize the remaining adaptability of a model in terms of compatible effective rank, defined as the log-volume of task-preserving directions that remain dynamically accessible. Although task performance may remain unchanged, finite-time learning can progressively reduce this reconfiguration capacity. We prove a capacity-threshold criterion for continual learning: let m_B denote the stable rank of the Hessian of a new task B restricted to the task-preserving manifold of a previously learned task A. If m_B exceeds the residual compatible effective rank, then task B is trajectory-level incompatible with task A; any sufficient adaptation necessarily induces forgetting. Thus catastrophic forgetting arises not from the absence of multi-task solutions, but from irreversible loss of reconfiguration capacity under compositional learning dynamics. This establishes a trajectory-level capacity limit for continual learning. --- ### IV Co-Scientist: Multi-Agent LLM Framework for Causal Instrumental Variable Discovery Authors: Ivaxi Sheth, Zhijing Jin, Bryan Wilder, Dominik Janzing, Mario Fritz Published: 2026-02-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.07943v2 In the presence of confounding between an endogenous variable and the outcome, instrumental variables (IVs) are used to isolate the causal effect of the endogenous variable. Identifying valid instruments requires interdisciplinary knowledge, creativity, and contextual understanding, making it a non-trivial task. In this paper, we investigate whether large language models (LLMs) can aid in this task. We perform a two-stage evaluation framework. First, we test whether LLMs can recover well-established instruments from the literature, assessing their ability to replicate standard reasoning. Second, we evaluate whether LLMs can identify and avoid instruments that have been empirically or theoretically discredited. Building on these results, we introduce IV Co-Scientist, a multi-agent system that proposes, critiques, and refines IVs for a given treatment-outcome pair. We also introduce a statistical test to contextualize consistency in the absence of ground truth. Our results show the potential of LLMs to discover valid instrumental variables from a large observational database. --- ### MePo: Meta Post-Refinement for Rehearsal-Free General Continual Learning Authors: Guanglong Sun, Hongwei Yan, Liyuan Wang, Zhiqi Kang, Shuang Cui, Hang Su, Jun Zhu, Yi Zhong Published: 2026-02-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.07940v3 To cope with uncertain changes of the external world, intelligent systems must continually learn from complex, evolving environments and respond in real time. This ability, collectively known as general continual learning (GCL), encapsulates practical challenges such as online datastreams and blurry task boundaries. Although leveraging pretrained models (PTMs) has greatly advanced conventional continual learning (CL), these methods remain limited in reconciling the diverse and temporally mixed information along a single pass, resulting in sub-optimal GCL performance. Inspired by meta-plasticity and reconstructive memory in neuroscience, we introduce here an innovative approach named Meta Post-Refinement (MePo) for PTMs-based GCL. This approach constructs pseudo task sequences from pretraining data and develops a bi-level meta-learning paradigm to refine the pretrained backbone, which serves as a prolonged pretraining phase but greatly facilitates rapid adaptation of representation learning to downstream GCL tasks. MePo further initializes a meta covariance matrix as the reference geometry of pretrained representation space, enabling GCL to exploit second-order statistics for robust output alignment. MePo serves as a plug-in strategy that achieves significant performance gains across a variety of GCL benchmarks and pretrained checkpoints in a rehearsal-free manner (e.g., 15.10\%, 13.36\%, and 12.56\% on CIFAR-100, ImageNet-R, and CUB-200 under Sup-21/1K). Our source code is available at \href{https://github.com/SunGL001/MePo}{MePo} --- ### Integrating Specialized and Generic Agent Motion Prediction with Dynamic Occupancy Grid Maps Authors: Rabbia Asghar, Lukas Rummelhard, Wenqian Liu, Anne Spalanzani, Christian Laugier Published: 2026-02-08 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2602.07938v1 Accurate prediction of driving scene is a challenging task due to uncertainty in sensor data, the complex behaviors of agents, and the possibility of multiple feasible futures. Existing prediction methods using occupancy grid maps primarily focus on agent-agnostic scene predictions, while agent-specific predictions provide specialized behavior insights with the help of semantic information. However, both paradigms face distinct limitations: agent-agnostic models struggle to capture the behavioral complexities of dynamic actors, whereas agent-specific approaches fail to generalize to poorly perceived or unrecognized agents; combining both enables robust and safer motion forecasting. To address this, we propose a unified framework by leveraging Dynamic Occupancy Grid Maps within a streamlined temporal decoding pipeline to simultaneously predict future occupancy state grids, vehicle grids, and scene flow grids. Relying on a lightweight spatiotemporal backbone, our approach is centered on a tailored, interdependent loss function that captures inter-grid dependencies and enables diverse future predictions. By using occupancy state information to enforce flow-guided transitions, the loss function acts as a regularizer that directs occupancy evolution while accounting for obstacles and occlusions. Consequently, the model not only predicts the specific behaviors of vehicle agents, but also identifies other dynamic entities and anticipates their evolution within the complex scene. Evaluations on real-world nuScenes and Woven Planet datasets demonstrate superior prediction performances for dynamic vehicles and generic dynamic scene elements compared to baseline methods. --- ### Attention-Based Deep Learning for Early Parkinson's Disease Detection with Tabular Biomedical Data Authors: Olamide Samuel Oseni, Ibraheem Omotolani Obanla, Toheeb Aduramomi Jimoh Published: 2026-02-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.07933v1 Early and accurate detection of Parkinson's disease (PD) remains a critical challenge in medical diagnostics due to the subtlety of early-stage symptoms and the complex, non-linear relationships inherent in biomedical data. Traditional machine learning (ML) models, though widely applied to PD detection, often rely on extensive feature engineering and struggle to capture complex feature interactions. This study investigates the effectiveness of attention-based deep learning models for early PD detection using tabular biomedical data. We present a comparative evaluation of four classification models: Multi-Layer Perceptron (MLP), Gradient Boosting, TabNet, and SAINT, using a benchmark dataset from the UCI Machine Learning Repository consisting of biomedical voice measurements from PD patients and healthy controls. Experimental results show that SAINT consistently outperformed all baseline models across multiple evaluation metrics, achieving a weighted precision of 0.98, weighted recall of 0.97, weighted F1-score of 0.97, a Matthews Correlation Coefficient (MCC) of 0.9990, and the highest Area Under the ROC Curve (AUC-ROC). TabNet and MLP demonstrated competitive performance, while Gradient Boosting yielded the lowest overall scores. The superior performance of SAINT is attributed to its dual attention mechanism, which effectively models feature interactions within and across samples. These findings demonstrate the diagnostic potential of attention-based deep learning architectures for early Parkinson's disease detection and highlight the importance of dynamic feature representation in clinical prediction tasks. --- ### Which private attributes do VLMs agree on and predict well? Authors: Olena Hrynenko, Darya Baranouskaya, Alina Elena Baia, Andrea Cavallaro Published: 2026-02-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.07931v1 Visual Language Models (VLMs) are often used for zero-shot detection of visual attributes in the image. We present a zero-shot evaluation of open-source VLMs for privacy-related attribute recognition. We identify the attributes for which VLMs exhibit strong inter-annotator agreement, and discuss the disagreement cases of human and VLM annotations. Our results show that when evaluated against human annotations, VLMs tend to predict the presence of privacy attributes more often than human annotators. In addition to this, we find that in cases of high inter-annotator agreement between VLMs, they can complement human annotation by identifying attributes overlooked by human annotators. This highlights the potential of VLMs to support privacy annotations in large-scale image datasets. --- ### Patches of Nonlinearity: Instruction Vectors in Large Language Models Authors: Irina Bigoulaeva, Jonas Rohweder, Subhabrata Dutta, Iryna Gurevych Published: 2026-02-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.07930v2 Despite the recent success of instruction-tuned language models and their ubiquitous usage, very little is known of how models process instructions internally. In this work, we address this gap from a mechanistic point of view by investigating how instruction-specific representations are constructed and utilized in different stages of post-training: Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO). Via causal mediation, we identify that instruction representation is fairly localized in models. These representations, which we call Instruction Vectors (IVs), demonstrate a curious juxtaposition of linear separability along with non-linear causal interaction, broadly questioning the scope of the linear representation hypothesis commonplace in mechanistic interpretability. To disentangle the non-linear causal interaction, we propose a novel method to localize information processing in language models that is free from the implicit linear assumptions of patching-based techniques. We find that, conditioned on the task representations formed in the early layers, different information pathways are selected in the later layers to solve that task, i.e., IVs act as circuit selectors. --- ### A Kinetic Energy Perspective of Flow Matching Authors: Ziyun Li, Huancheng Hu, Soon Hoe Lim, Xuyu Li, Fei Gao, Enmao Diao, Zezhen Ding, Michalis Vazirgiannis, Henrik Bostrom Published: 2026-02-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.07928v2 Flow-based generative models can be viewed through a physics lens: sampling transports a particle from noise to data by integrating a learned velocity field, and each sample corresponds to a trajectory with its own dynamical effort. Motivated by classical mechanics, we introduce Kinetic Path Energy (KPE), an action-like, per-sample diagnostic that measures the accumulated kinetic effort along an ordinary differential equation (ODE) trajectory. Empirically, KPE exhibits two robust correspondences: {i} higher KPE predicts stronger semantic fidelity; {ii} high-KPE trajectories land in sparse representation regions. We further provide theoretical guarantees linking trajectory energy to data sparsity. Paradoxically, this correlation is non-monotonic. At sufficiently high energy, generation can degenerate into memorization. Leveraging the closed-form formula of empirical flow matching, we show that extreme energies drive trajectories toward near-copies of training examples. This yields a Goldilocks principle and motivates Kinetic Trajectory Shaping (KTS), a training-free two-phase inference strategy that boosts early motion and enforces a late-time soft landing, reducing memorization and improving generation quality across benchmark tasks. --- ### Optimized Human-Robot Co-Dispatch Planning for Petro-Site Surveillance under Varying Criticalities Authors: Nur Ahmad Khatim, Mansur Arief Published: 2026-02-08 Categories: cs.RO, cs.AI, math.OC arXiv: https://arxiv.org/abs/2602.07924v2 Securing petroleum infrastructure requires balancing autonomous system efficiency with human judgment for threat escalation, a challenge unaddressed by classical facility location models assuming homogeneous resources. This paper formulates the Human-Robot Co-Dispatch Facility Location Problem (HRCD-FLP), a capacitated facility location variant incorporating tiered infrastructure criticality, human-robot supervision ratio constraints, and minimum utilization requirements. We evaluate command center selection across three technology maturity scenarios. Results show transitioning from conservative (1:3 human-robot supervision) to future autonomous operations (1:10) yields significant cost reduction while maintaining complete critical infrastructure coverage. For small problems, exact methods dominate in both cost and computation time; for larger problems, the proposed heuristic achieves feasible solutions in under 3 minutes with approximately 14% optimality gap where comparison is possible. From systems perspective, our work demonstrate that optimized planning for human-robot teaming is key to achieve both cost-effective and mission-reliable deployments. --- ### Selective Fine-Tuning for Targeted and Robust Concept Unlearning Authors: Mansi, Avinash Kori, Francesca Toni, Soteris Demetriou Published: 2026-02-08 Categories: cs.AI, cs.CV arXiv: https://arxiv.org/abs/2602.07919v2 Text guided diffusion models are used by millions of users, but can be easily exploited to produce harmful content. Concept unlearning methods aim at reducing the models' likelihood of generating harmful content. Traditionally, this has been tackled at an individual concept level, with only a handful of recent works considering more realistic concept combinations. However, state of the art methods depend on full finetuning, which is computationally expensive. Concept localisation methods can facilitate selective finetuning, but existing techniques are static, resulting in suboptimal utility. In order to tackle these challenges, we propose TRUST (Targeted Robust Selective fine Tuning), a novel approach for dynamically estimating target concept neurons and unlearning them through selective finetuning, empowered by a Hessian based regularization. We show experimentally, against a number of SOTA baselines, that TRUST is robust against adversarial prompts, preserves generation quality to a significant degree, and is also significantly faster than the SOTA. Our method achieves unlearning of not only individual concepts but also combinations of concepts and conditional concepts, without any specific regularization. --- ### Selective Fine-Tuning for Targeted and Robust Concept Unlearning Authors: Mansi, Avinash Kori, Francesca Toni, Soteris Demetriou Published: 2026-02-08 Categories: cs.AI, cs.CV arXiv: https://arxiv.org/abs/2602.07919v1 Text guided diffusion models are used by millions of users, but can be easily exploited to produce harmful content. Concept unlearning methods aim at reducing the models' likelihood of generating harmful content. Traditionally, this has been tackled at an individual concept level, with only a handful of recent works considering more realistic concept combinations. However, state of the art methods depend on full finetuning, which is computationally expensive. Concept localisation methods can facilitate selective finetuning, but existing techniques are static, resulting in suboptimal utility. In order to tackle these challenges, we propose TRUST (Targeted Robust Selective fine Tuning), a novel approach for dynamically estimating target concept neurons and unlearning them through selective finetuning, empowered by a Hessian based regularization. We show experimentally, against a number of SOTA baselines, that TRUST is robust against adversarial prompts, preserves generation quality to a significant degree, and is also significantly faster than the SOTA. Our method achieves unlearning of not only individual concepts but also combinations of concepts and conditional concepts, without any specific regularization. --- ### CausalArmor: Efficient Indirect Prompt Injection Guardrails via Causal Attribution Authors: Minbeom Kim, Mihir Parmar, Phillip Wallis, Lesly Miculicich, Kyomin Jung, Krishnamurthy Dj Dvijotham, Long T. Le, Tomas Pfister Published: 2026-02-08 Categories: cs.CR, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2602.07918v1 AI agents equipped with tool-calling capabilities are susceptible to Indirect Prompt Injection (IPI) attacks. In this attack scenario, malicious commands hidden within untrusted content trick the agent into performing unauthorized actions. Existing defenses can reduce attack success but often suffer from the over-defense dilemma: they deploy expensive, always-on sanitization regardless of actual threat, thereby degrading utility and latency even in benign scenarios. We revisit IPI through a causal ablation perspective: a successful injection manifests as a dominance shift where the user request no longer provides decisive support for the agent's privileged action, while a particular untrusted segment, such as a retrieved document or tool output, provides disproportionate attributable influence. Based on this signature, we propose CausalArmor, a selective defense framework that (i) computes lightweight, leave-one-out ablation-based attributions at privileged decision points, and (ii) triggers targeted sanitization only when an untrusted segment dominates the user intent. Additionally, CausalArmor employs retroactive Chain-of-Thought masking to prevent the agent from acting on ``poisoned'' reasoning traces. We present a theoretical analysis showing that sanitization based on attribution margins conditionally yields an exponentially small upper bound on the probability of selecting malicious actions. Experiments on AgentDojo and DoomArena demonstrate that CausalArmor matches the security of aggressive defenses while improving explainability and preserving utility and latency of AI agents. --- ### CausalCompass: Evaluating the Robustness of Time-Series Causal Discovery in Misspecified Scenarios Authors: Huiyang Yi, Xiaojian Shen, Yonggang Wu, Duxin Chen, He Wang, Wenwu Yu Published: 2026-02-08 Categories: cs.LG, cs.AI, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2602.07915v2 Causal discovery from time series is a fundamental task in machine learning. However, its widespread adoption is hindered by a reliance on untestable causal assumptions and by the lack of robustness-oriented evaluation in existing benchmarks. To address these challenges, we propose CausalCompass, a flexible and extensible benchmark framework designed to assess the robustness of time-series causal discovery (TSCD) methods under violations of modeling assumptions. To demonstrate the practical utility of CausalCompass, we conduct extensive benchmarking of representative TSCD algorithms across eight assumption-violation scenarios. Our experimental results indicate that no single method consistently attains optimal performance across all settings. Nevertheless, the methods exhibiting superior overall performance across diverse scenarios are almost invariably deep learning-based approaches. We further provide hyperparameter sensitivity analyses to deepen the understanding of these findings. We additionally conduct ablation experiments to explain the strong performance of deep learning-based methods under assumption violations. We also find, somewhat surprisingly, that NTS-NOTEARS relies heavily on standardized preprocessing in practice, performing poorly in the vanilla setting but exhibiting strong performance after standardization. Finally, our work aims to provide a comprehensive and systematic evaluation of TSCD methods under assumption violations, thereby facilitating their broader adoption in real-world applications. The user-friendly implementation, documentation and datasets are available at https://anonymous.4open.science/r/CausalCompass-anonymous-5B4F/. --- ### SparseEval: Efficient Evaluation of Large Language Models by Sparse Optimization Authors: Taolin Zhang, Hang Guo, Wang Lu, Tao Dai, Shu-Tao Xia, Jindong Wang Published: 2026-02-08 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.07909v1 As large language models (LLMs) continue to scale up, their performance on various downstream tasks has significantly improved. However, evaluating their capabilities has become increasingly expensive, as performing inference on a large number of benchmark samples incurs high computational costs. In this paper, we revisit the model-item performance matrix and show that it exhibits sparsity, that representative items can be selected as anchors, and that the task of efficient benchmarking can be formulated as a sparse optimization problem. Based on these insights, we propose SparseEval, a method that, for the first time, adopts gradient descent to optimize anchor weights and employs an iterative refinement strategy for anchor selection. We utilize the representation capacity of MLP to handle sparse optimization and propose the Anchor Importance Score and Candidate Importance Score to evaluate the value of each item for task-aware refinement. Extensive experiments demonstrate the low estimation error and high Kendall's~$τ$ of our method across a variety of benchmarks, showcasing its superior robustness and practicality in real-world scenarios. Code is available at {https://github.com/taolinzhang/SparseEval}. ---