# Research Papers — 2026-02 · bundle page 207 of 210 > Full-text markdown of papers #10301–10350 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/207 [← Previous bundle](https://aipod.dev/papers/archive/2026-02/page/206.md) [Next bundle →](https://aipod.dev/papers/archive/2026-02/page/208.md) ### Seeing, Hearing, and Knowing Together: Multimodal Strategies in Deepfake Videos Detection Authors: Chen Chen, Dion Hoe-Lian Goh Published: 2026-02-01 Categories: cs.MM, cs.CV, cs.HC arXiv: https://arxiv.org/abs/2602.01284v1 As deepfake videos become increasingly difficult for people to recognise, understanding the strategies humans use is key to designing effective media literacy interventions. We conducted a study with 195 participants between the ages of 21 and 40, who judged real and deepfake videos, rated their confidence, and reported the cues they relied on across visual, audio, and knowledge strategies. Participants were more accurate with real videos than with deepfakes and showed lower expected calibration error for real content. Through association rule mining, we identified cue combinations that shaped performance. Visual appearance, vocal, and intuition often co-occurred for successful identifications, which highlights the importance of multimodal approaches in human detection. Our findings show which cues help or hinder detection and suggest directions for designing media literacy tools that guide effective cue use. Building on these insights can help people improve their identification skills and become more resilient to deceptive digital media. --- ### Who Transfers Safety? Identifying and Targeting Cross-Lingual Shared Safety Neurons Authors: Xianhui Zhang, Chengyu Xie, Linxia Zhu, Yonghui Yang, Weixiang Zhao, Zifeng Cheng, Cong Wang, Fei Shen, Tat-Seng Chua Published: 2026-02-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01283v1 Multilingual safety remains significantly imbalanced, leaving non-high-resource (NHR) languages vulnerable compared to robust high-resource (HR) ones. Moreover, the neural mechanisms driving safety alignment remain unclear despite observed cross-lingual representation transfer. In this paper, we find that LLMs contain a set of cross-lingual shared safety neurons (SS-Neurons), a remarkably small yet critical neuronal subset that jointly regulates safety behavior across languages. We first identify monolingual safety neurons (MS-Neurons) and validate their causal role in safety refusal behavior through targeted activation and suppression. Our cross-lingual analyses then identify SS-Neurons as the subset of MS-Neurons shared between HR and NHR languages, serving as a bridge to transfer safety capabilities from HR to NHR domains. We observe that suppressing these neurons causes concurrent safety drops across NHR languages, whereas reinforcing them improves cross-lingual defensive consistency. Building on these insights, we propose a simple neuron-oriented training strategy that targets SS-Neurons based on language resource distribution and model architecture. Experiments demonstrate that fine-tuning this tiny neuronal subset outperforms state-of-the-art methods, significantly enhancing NHR safety while maintaining the model's general capabilities. The code and dataset will be available athttps://github.com/1518630367/SS-Neuron-Expansion. --- ### Richer Bayesian Last Layers with Subsampled NTK Features Authors: Sergio Calvo-Ordoñez, Jonathan Plenk, Richard Bergna, Álvaro Cartea, Yarin Gal, Jose Miguel Hernández-Lobato, Kamil Ciosek Published: 2026-02-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01279v2 Bayesian Last Layers (BLLs) provide a convenient and computationally efficient way to estimate uncertainty in neural networks. However, they underestimate epistemic uncertainty because they apply a Bayesian treatment only to the final layer, ignoring uncertainty induced by earlier layers. We propose a method that improves BLLs by leveraging a projection of Neural Tangent Kernel (NTK) features onto the space spanned by the last-layer features. This enables posterior inference that accounts for variability of the full network while retaining the low computational cost of inference of a standard BLL. We show that our method yields posterior variances that are provably greater or equal to those of a standard BLL, correcting its tendency to underestimate epistemic uncertainty. To further reduce computational cost, we introduce a uniform subsampling scheme for estimating the projection matrix and for posterior inference. We derive approximation bounds for both types of subsampling. Empirical evaluations on UCI regression, contextual bandits, image classification, and out-of-distribution detection tasks in image and tabular datasets, demonstrate improved calibration and uncertainty estimates compared to standard BLLs and competitive baselines, while reducing computational cost. --- ### DSFC-Net: A Dual-Encoder Spatial and Frequency Co-Awareness Network for Rural Road Extraction Authors: Zhengbo Zhang, Yihe Tian, Wanke Xia, Lin Chen, Yue Sun, Kun Ding, Ying Wang, Bing Xu, Shiming Xiang Published: 2026-02-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01278v1 Accurate extraction of rural roads from high-resolution remote sensing imagery is essential for infrastructure planning and sustainable development. However, this task presents unique challenges in rural settings due to several factors. These include high intra-class variability and low inter-class separability from diverse surface materials, frequent vegetation occlusions that disrupt spatial continuity, and narrow road widths that exacerbate detection difficulties. Existing methods, primarily optimized for structured urban environments, often underperform in these scenarios as they overlook such distinctive characteristics. To address these challenges, we propose DSFC-Net, a dual-encoder framework that synergistically fuses spatial and frequency-domain information. Specifically, a CNN branch is employed to capture fine-grained local road boundaries and short-range continuity, while a novel Spatial-Frequency Hybrid Transformer (SFT) is introduced to robustly model global topological dependencies against vegetation occlusions. Distinct from standard attention mechanisms that suffer from frequency bias, the SFT incorporates a Cross-Frequency Interaction Attention (CFIA) module that explicitly decouples high- and low-frequency information via a Laplacian Pyramid strategy. This design enables the dynamic interaction between spatial details and frequency-aware global contexts, effectively preserving the connectivity of narrow roads. Furthermore, a Channel Feature Fusion Module (CFFM) is proposed to bridge the two branches by adaptively recalibrating channel-wise feature responses, seamlessly integrating local textures with global semantics for accurate segmentation. Comprehensive experiments on the WHU-RuR+, DeepGlobe, and Massachusetts datasets validate the superiority of DSFC-Net over state-of-the-art approaches. --- ### TF-Lane: Traffic Flow Module for Robust Lane Perception Authors: Yihan Xie, Han Xia, Zhen Yang Published: 2026-02-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01277v1 Autonomous driving systems require robust lane perception capabilities, yet existing vision-based detection methods suffer significant performance degradation when visual sensors provide insufficient cues, such as in occluded or lane-missing scenarios. While some approaches incorporate high-definition maps as supplementary information, these solutions face challenges of high subscription costs and limited real-time performance. To address these limitations, we explore an innovative information source: traffic flow, which offers real-time capabilities without additional costs. This paper proposes a TrafficFlow-aware Lane perception Module (TFM) that effectively extracts real-time traffic flow features and seamlessly integrates them with existing lane perception algorithms. This solution originated from real-world autonomous driving conditions and was subsequently validated on open-source algorithms and datasets. Extensive experiments on four mainstream models and two public datasets (Nuscenes and OpenLaneV2) using standard evaluation metrics show that TFM consistently improves performance, achieving up to +4.1% mAP gain on the Nuscenes dataset. --- ### LLM-Driven Ontology Construction for Enterprise Knowledge Graphs Authors: Abdulsobur Oyewale, Tommaso Soru Published: 2026-02-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01276v1 Enterprise Knowledge Graphs have become essential for unifying heterogeneous data and enforcing semantic governance. However, the construction of their underlying ontologies remains a resource-intensive, manual process that relies heavily on domain expertise. This paper introduces OntoEKG, a LLM-driven pipeline designed to accelerate the generation of domain-specific ontologies from unstructured enterprise data. Our approach decomposes the modelling task into two distinct phases: an extraction module that identifies core classes and properties, and an entailment module that logically structures these elements into a hierarchy before serialising them into standard RDF. Addressing the significant lack of comprehensive benchmarks for end-to-end ontology construction, we adopt a new evaluation dataset derived from documents across the Data, Finance, and Logistics sectors. Experimental results highlight both the potential and the challenges of this approach, achieving a fuzzy-match F1-score of 0.724 in the Data domain while revealing limitations in scope definition and hierarchical reasoning. --- ### PACER: Blockwise Pre-verification for Speculative Decoding with Adaptive Length Authors: Situo Zhang, Yifan Zhang, Zichen Zhu, Hankun Wang, Da Ma, Danyang Zhang, Lu Chen, Kai Yu Published: 2026-02-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.01274v1 Speculative decoding (SD) is a powerful technique for accelerating the inference process of large language models (LLMs) without sacrificing accuracy. Typically, SD employs a small draft model to generate a fixed number of draft tokens, which are then verified in parallel by the target model. However, our experiments reveal that the optimal draft length varies significantly across different decoding steps. This variation suggests that using a fixed draft length limits the potential for further improvements in decoding speed. To address this challenge, we propose Pacer, a novel approach that dynamically controls draft length using a lightweight, trainable pre-verification layer. This layer pre-verifies draft tokens blockwise before they are sent to the target model, allowing the draft model to stop token generation if the blockwise pre-verification fails. We implement Pacer on multiple SD model pairs and evaluate its performance across various benchmarks. Our results demonstrate that Pacer achieves up to 2.66x Speedup over autoregressive decoding and consistently outperforms standard speculative decoding. Furthermore, when integrated with Ouroboros, Pacer attains up to 3.09x Speedup. --- ### Q-DiT4SR: Exploration of Detail-Preserving Diffusion Transformer Quantization for Real-World Image Super-Resolution Authors: Xun Zhang, Kaicheng Yang, Hongliang Lu, Haotong Qin, Yong Guo, Yulun Zhang Published: 2026-02-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01273v4 Recently, Diffusion Transformers (DiTs) have emerged in Real-World Image Super-Resolution (Real-ISR) to generate high-quality textures, yet their heavy inference burden hinders real-world deployment. While Post-Training Quantization (PTQ) is a promising solution for acceleration, existing methods in super-resolution mostly focus on U-Net architectures, whereas generic DiT quantization is typically designed for text-to-image tasks. Directly applying these methods to DiT-based super-resolution models leads to severe degradation of local textures. Therefore, we propose Q-DiT4SR, the first PTQ framework specifically tailored for DiT-based Real-ISR. We propose H-SVD, a hierarchical SVD that integrates a global low-rank branch with a local block-wise rank-1 branch under a matched parameter budget. We further propose Variance-aware Spatio-Temporal Mixed Precision: VaSMP allocates cross-layer weight bit-widths in a data-free manner based on rate-distortion theory, while VaTMP schedules intra-layer activation precision across diffusion timesteps via dynamic programming (DP) with minimal calibration. Experiments on multiple real-world datasets demonstrate that our Q-DiT4SR achieves SOTA performance under both W4A6 and W4A4 settings. Notably, the W4A4 quantization configuration reduces model size by 5.8$\times$ and computational operations by 6.14$\times$. Our code and models will be available at https://github.com/xunzhang1128/Q-DiT4SR. --- ### From Intents to Actions: Agentic AI in Autonomous Networks Authors: Burak Demirel, Pablo Soldati, Yu Wang Published: 2026-02-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01271v1 Telecommunication networks are increasingly expected to operate autonomously while supporting heterogeneous services with diverse and often conflicting intents -- that is, performance objectives, constraints, and requirements specific to each service. However, transforming high-level intents -- such as ultra-low latency, high throughput, or energy efficiency -- into concrete control actions (i.e., low-level actuator commands) remains beyond the capability of existing heuristic approaches. This work introduces an Agentic AI system for intent-driven autonomous networks, structured around three specialized agents. A supervisory interpreter agent, powered by language models, performs both lexical parsing of intents into executable optimization templates and cognitive refinement based on feedback, constraint feasibility, and evolving network conditions. An optimizer agent converts these templates into tractable optimization problems, analyzes trade-offs, and derives preferences across objectives. Lastly, a preference-driven controller agent, based on multi-objective reinforcement learning, leverages these preferences to operate near the Pareto frontier of network performance that best satisfies the original intent. Collectively, these agents enable networks to autonomously interpret, reason over, adapt to, and act upon diverse intents and network conditions in a scalable manner. --- ### Mixture-of-World Models: Scaling Multi-Task Reinforcement Learning with Modular Latent Dynamics Authors: Boxuan Zhang, Weipu Zhang, Zhaohan Feng, Wei Xiao, Jian Sun, Jie Chen, Gang Wang Published: 2026-02-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01270v1 A fundamental challenge in multi-task reinforcement learning (MTRL) is achieving sample efficiency in visual domains where tasks exhibit substantial heterogeneity in both observations and dynamics. Model-based reinforcement learning offers a promising path to improved sample efficiency through world models, but standard monolithic architectures struggle to capture diverse task dynamics, resulting in poor reconstruction and prediction accuracy. We introduce Mixture-of-World Models (MoW), a scalable architecture that combines modular variational autoencoders for task-adaptive visual compression, a hybrid Transformer-based dynamics model with task-conditioned experts and a shared backbone, and a gradient-based task clustering strategy for efficient parameter allocation. On the Atari 100k benchmark, a single MoW agent trained once on 26 Atari games achieves a mean human-normalized score of 110.4%, competitive with the score of 114.2% achieved by STORM, an ensemble of 26 task-specific models, while using 50% fewer parameters. On Meta-World, MoW achieves a 74.5% average success rate within 300 thousand environment steps, establishing a new state of the art. These results demonstrate that MoW provides a scalable and parameter-efficient foundation for generalist world models. --- ### OASIS-DC: Generalizable Depth Completion via Output-level Alignment of Sparse-Integrated Monocular Pseudo Depth Authors: Jaehyeon Cho, Jhonghyun An Published: 2026-02-01 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2602.01268v1 Recent monocular foundation models excel at zero-shot depth estimation, yet their outputs are inherently relative rather than metric, limiting direct use in robotics and autonomous driving. We leverage the fact that relative depth preserves global layout and boundaries: by calibrating it with sparse range measurements, we transform it into a pseudo metric depth prior. Building on this prior, we design a refinement network that follows the prior where reliable and deviates where necessary, enabling accurate metric predictions from very few labeled samples. The resulting system is particularly effective when curated validation data are unavailable, sustaining stable scale and sharp edges across few-shot regimes. These findings suggest that coupling foundation priors with sparse anchors is a practical route to robust, deployment-ready depth completion under real-world label scarcity. --- ### Diving into Kronecker Adapters: Component Design Matters Authors: Jiayu Bai, Danchen Yu, Zhenyu Liao, TianQi Hou, Feng Zhou, Robert C. Qiu, Zenan Ling Published: 2026-02-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01267v3 Kronecker adapters have emerged as a promising approach for fine-tuning large-scale models, enabling high-rank updates through tunable component structures. However, existing work largely treats the component structure as a fixed or heuristic design choice, leaving the dimensions and number of Kronecker components underexplored. In this paper, we identify component structure as a key factor governing the capacity of Kronecker adapters. We perform a fine-grained analysis of both the dimensions and number of Kronecker components. In particular, we show that the alignment between Kronecker adapters and full fine-tuning depends on component configurations. Guided by these insights, we propose Component Designed Kronecker Adapters (CDKA). We further provide parameter-budget-aware configuration guidelines and a tailored training stabilization strategy for practical deployment. Experiments across various architectures and modalities demonstrate the effectiveness of CDKA. Code is available at https://github.com/rainstonee/CDKA. --- ### Diving into Kronecker Adapters: Component Design Matters Authors: Jiayu Bai, Danchen Yu, Zhenyu Liao, TianQi Hou, Feng Zhou, Robert C. Qiu, Zenan Ling Published: 2026-02-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01267v2 Kronecker adapters have emerged as a promising approach for fine-tuning large-scale models, enabling high-rank updates through tunable component structures. However, existing work largely treats the component structure as a fixed or heuristic design choice, leaving the dimensions and number of Kronecker components underexplored. In this paper, we identify component structure as a key factor governing the capacity of Kronecker adapters. We perform a fine-grained analysis of both the dimensions and number of Kronecker components. In particular, we show that the alignment between Kronecker adapters and full fine-tuning depends on component configurations. Guided by these insights, we propose Component Designed Kronecker Adapters (CDKA). We further provide parameter-budget-aware configuration guidelines and a tailored training stabilization strategy for practical deployment. Experiments across various architectures and modalities demonstrate the effectiveness of CDKA. Code is available at https://github.com/rainstonee/CDKA. --- ### BicKD: Bilateral Contrastive Knowledge Distillation Authors: Jiangnan Zhu, Yukai Xu, Li Xiong, Yixuan Liu, Junxu Liu, Hong kyu Lee, Yujie Gu Published: 2026-02-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01265v2 Knowledge distillation (KD) is a machine learning framework that transfers knowledge from a teacher model to a student model. The vanilla KD proposed by Hinton et al. has been the dominant approach in logit-based distillation and demonstrates compelling performance. However, it only performs sample-wise probability alignment between teacher and student's predictions, lacking an mechanism for class-wise comparison. Besides, vanilla KD imposes no structural constraint on the probability space. In this work, we propose a simple yet effective methodology, bilateral contrastive knowledge distillation (BicKD). This approach introduces a novel bilateral contrastive loss, which intensifies the orthogonality among different class generalization spaces while preserving consistency within the same class. The bilateral formulation enables explicit comparison of both sample-wise and class-wise prediction patterns between teacher and student. By emphasizing probabilistic orthogonality, BicKD further regularizes the geometric structure of the predictive distribution. Extensive experiments show that our BicKD method enhances knowledge transfer, and consistently outperforms state-of-the-art knowledge distillation techniques across various model architectures and benchmarks. --- ### Sample Efficient Active Algorithms for Offline Reinforcement Learning Authors: Soumyadeep Roy, Shashwat Kushwaha, Ambedkar Dukkipati Published: 2026-02-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.01260v1 Offline reinforcement learning (RL) enables policy learning from static data but often suffers from poor coverage of the state-action space and distributional shift problems. This problem can be addressed by allowing limited online interactions to selectively refine uncertain regions of the learned value function, which is referred to as Active Reinforcement Learning (ActiveRL). While there has been good empirical success, no theoretical analysis is available in the literature. We fill this gap by developing a rigorous sample-complexity analysis of ActiveRL through the lens of Gaussian Process (GP) uncertainty modeling. In this respect, we propose an algorithm and using GP concentration inequalities and information-gain bounds, we derive high-probability guarantees showing that an $ε$-optimal policy can be learned with ${\mathcal{O}}(1/ε^2)$ active transitions, improving upon the $Ω(1/ε^2(1-γ)^4)$ rate of purely offline methods. Our results reveal that ActiveRL achieves near-optimal information efficiency, that is, guided uncertainty reduction leads to accelerated value-function convergence with minimal online data. Our analysis builds on GP concentration inequalities and information-gain bounds, bridging Bayesian nonparametric regression and reinforcement learning theories. We conduct several experiments to validate the algorithm and theoretical findings. --- ### Boosting Point-supervised Temporal Action Localization via Text Refinement and Alignment Authors: Yunchuan Ma, Laiyun Qing, Guorong Li, Yuqing Liu, Yuankai Qi, Qingming Huang Published: 2026-02-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01257v2 Recently, point-supervised temporal action localization has gained significant attention for its effective balance between labeling costs and localization accuracy. However, current methods primarily rely on visual features and do not fully exploit the complementary semantic information contained in textual descriptions. To address this issue, we propose a Text Refinement and Alignment (TRA) framework that incorporates refined textual semantics to complement visual representations for point-supervised temporal action localization. This is achieved by designing two new modules for the original point-supervised framework: a Point-based Text Refinement module (PTR) and a Point-based Multimodal Alignment module (PMA). Specifically, we first generate descriptions for video frames using a pre-trained multimodal model. Next, PTR refines the initial descriptions by leveraging point annotations together with multiple fixed pre-trained models. PMA then projects the visual and textual features into a unified semantic space and employs a point-based multimodal contrastive learning objective to reduce the gap between visual and linguistic modalities. Finally, the aligned multimodal features are fed into the action detector for temporal action localization. Extensive experiments on five widely used benchmarks demonstrate that TRA consistently improves strong point-supervised baselines and achieves competitive performance compared with state-of-the-art methods. In particular, on THUMOS'14, TRA achieves 58.5% AVG mAP@(0.1:0.7), outperforming the visual-only baseline by 2.3%. --- ### Boosting Point-supervised Temporal Action Localization via Text Refinement and Alignment Authors: Yunchuan Ma, Laiyun Qing, Guorong Li, Yuqing Liu, Yuankai Qi, Qingming Huang Published: 2026-02-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01257v1 Recently, point-supervised temporal action localization has gained significant attention for its effective balance between labeling costs and localization accuracy. However, current methods only consider features from visual inputs, neglecting helpful semantic information from the text side. To address this issue, we propose a Text Refinement and Alignment (TRA) framework that effectively utilizes textual features from visual descriptions to complement the visual features as they are semantically rich. This is achieved by designing two new modules for the original point-supervised framework: a Point-based Text Refinement module (PTR) and a Point-based Multimodal Alignment module (PMA). Specifically, we first generate descriptions for video frames using a pre-trained multimodal model. Next, PTR refines the initial descriptions by leveraging point annotations together with multiple pre-trained models. PMA then projects all features into a unified semantic space and leverages a point-level multimodal feature contrastive learning to reduce the gap between visual and linguistic modalities. Last, the enhanced multi-modal features are fed into the action detector for precise localization. Extensive experimental results on five widely used benchmarks demonstrate the favorable performance of our proposed framework compared to several state-of-the-art methods. Moreover, our computational overhead analysis shows that the framework can run on a single 24 GB RTX 3090 GPU, indicating its practicality and scalability. --- ### Mechanistic Interpretability of Brain-to-Speech Models Across Speech Modes Authors: Maryam Maghsoudi, Ayushi Mishra Published: 2026-02-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.01247v1 Brain-to-speech decoding models demonstrate robust performance in vocalized, mimed, and imagined speech; yet, the fundamental mechanisms via which these models capture and transmit information across different speech modalities are less explored. In this work, we use mechanistic interpretability to causally investigate the internal representations of a neural speech decoder. We perform cross-mode activation patching of internal activations across speech modes, and use tri-modal interpolation to examine whether speech representations vary discretely or continuously. We use coarse-to-fine causal tracing and causal scrubbing to find localized causal structure, allowing us to find internal subspaces that are sufficient for cross-mode transfer. In order to determine how finely distributed these effects are within layers, we perform neuron-level activation patching. We discover that small but not distributed subsets of neurons, rather than isolated units, affect the cross-mode transfer. Our results show that speech modes lie on a shared continuous causal manifold, and cross-mode transfer is mediated by compact, layer-specific subspaces rather than diffuse activity. Together, our findings give a causal explanation for how speech modality information is organized and used in brain-to-speech decoding models, revealing hierarchical and direction-dependent representational structure across speech modes. --- ### PARSE: An Open-Domain Reasoning Question Answering Benchmark for Persian Authors: Jamshid Mozafari, Seyed Parsa Mousavinasab, Adam Jatowt Published: 2026-02-01 Categories: cs.CL, cs.IR arXiv: https://arxiv.org/abs/2602.01246v1 Reasoning-focused Question Answering (QA) has advanced rapidly with Large Language Models (LLMs), yet high-quality benchmarks for low-resource languages remain scarce. Persian, spoken by roughly 130 million people, lacks a comprehensive open-domain resource for evaluating reasoning-capable QA systems. We introduce PARSE, the first open-domain Persian reasoning QA benchmark, containing 10,800 questions across Boolean, multiple-choice, and factoid formats, with diverse reasoning types, difficulty levels, and answer structures. The benchmark is built via a controlled LLM-based generation pipeline and validated through human evaluation. We also ensure linguistic and factual quality through multi-stage filtering, annotation, and consistency checks. We benchmark multilingual and Persian LLMs under multiple prompting strategies and show that Persian prompts and structured prompting (CoT for Boolean/multiple-choice; few-shot for factoid) improve performance. Fine-tuning further boosts results, especially for Persian-specialized models. These findings highlight how PARSE supports both fair comparison and practical model adaptation. PARSE fills a critical gap in Persian QA research and provides a strong foundation for developing and evaluating reasoning-capable LLMs in low-resource settings. --- ### Large-Scale Terminal Agentic Trajectory Generation from Dockerized Environments Authors: Siwei Wu, Yizhi Li, Yuyang Song, Wei Zhang, Yang Wang, Riza Batista-Navarro, Xian Yang, Mingjie Tang, Bryan Dai, Jian Yang, Chenghua Lin Published: 2026-02-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.01244v3 Training agentic models for terminal-based tasks critically depends on high-quality terminal trajectories that capture realistic long-horizon interactions across diverse domains. However, constructing such data at scale remains challenging due to two key requirements: \textbf{\emph{Executability}}, since each instance requires a suitable and often distinct Docker environment; and \textbf{\emph{Verifiability}}, because heterogeneous task outputs preclude unified, standardized verification. To address these challenges, we propose \textbf{TerminalTraj}, a scalable pipeline that (i) filters high-quality repositories to construct Dockerized execution environments, (ii) generates Docker-aligned task instances, and (iii) synthesizes agent trajectories with executable validation code. Using TerminalTraj, we curate 32K Docker images and generate 50,733 verified terminal trajectories across eight domains. Models trained on this data with the Qwen2.5-Coder backbone achieve consistent performance improvements on TerminalBench (TB), with gains of up to 20\% on TB~1.0 and 10\% on TB~2.0 over their respective backbones. Notably, \textbf{TerminalTraj-32B} achieves strong performance among models with fewer than 100B parameters, reaching 35.30\% on TB~1.0 and 22.00\% on TB~2.0, and demonstrates improved test-time scaling behavior. All code and data are available at https://github.com/Wusiwei0410/TerminalTraj. --- ### Minimizing Mismatch Risk: A Prototype-Based Routing Framework for Zero-shot LLM-generated Text Detection Authors: Ke Sun, Guangsheng Bao, Han Cui, Yue Zhang Published: 2026-02-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.01240v1 Zero-shot methods detect LLM-generated text by computing statistical signatures using a surrogate model. Existing approaches typically employ a fixed surrogate for all inputs regardless of the unknown source. We systematically examine this design and find that detection performance varies substantially depending on surrogate-source alignment. We observe that while no single surrogate achieves optimal performance universally, a well-matched surrogate typically exists within a diverse pool for any given input. This finding transforms robust detection into a routing problem: selecting the most appropriate surrogate for each input. We propose DetectRouter, a prototype-based framework that learns text-detector affinity through two-stage training. The first stage constructs discriminative prototypes from white-box models; the second generalizes to black-box sources by aligning geometric distances with observed detection scores. Experiments on EvoBench and MAGE benchmarks demonstrate consistent improvements across multiple detection criteria and model families. --- ### Inferential Question Answering Authors: Jamshid Mozafari, Hamed Zamani, Guido Zuccon, Adam Jatowt Published: 2026-02-01 Categories: cs.CL, cs.IR arXiv: https://arxiv.org/abs/2602.01239v1 Despite extensive research on a wide range of question answering (QA) systems, most existing work focuses on answer containment-i.e., assuming that answers can be directly extracted and/or generated from documents in the corpus. However, some questions require inference, i.e., deriving answers that are not explicitly stated but can be inferred from the available information. We introduce Inferential QA -- a new task that challenges models to infer answers from answer-supporting passages which provide only clues. To study this problem, we construct QUIT (QUestions requiring Inference from Texts) dataset, comprising 7,401 questions and 2.4M passages built from high-convergence human- and machine-authored hints, labeled across three relevance levels using LLM-based answerability and human verification. Through comprehensive evaluation of retrievers, rerankers, and LLM-based readers, we show that methods effective on traditional QA tasks struggle in inferential QA: retrievers underperform, rerankers offer limited gains, and fine-tuning provides inconsistent improvements. Even reasoning-oriented LLMs fail to outperform smaller general-purpose models. These findings reveal that current QA pipelines are not yet ready for inference-based reasoning. Inferential QA thus establishes a new class of QA tasks that move towards understanding and reasoning from indirect textual evidence. --- ### Predictive Scheduling for Efficient Inference-Time Reasoning in Large Language Models Authors: Katrina Brown, Aneesh Muppidi, Rana Shahout Published: 2026-02-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01237v1 Large language models (LLMs) achieve state-of-the-art accuracy on complex reasoning tasks by generating multiple chain-of-thought (CoT) traces, but using a fixed token budget per query leads to over-computation on easy inputs and under-computation on hard ones. We introduce Predictive Scheduling, a plug-and-play framework that pre-runs lightweight predictors, an MLP on intermediate transformer hidden states or a LoRA-fine-tuned classifier on raw question text, to estimate each query's optimal reasoning length or difficulty before any full generation. Our greedy batch allocator dynamically distributes a fixed total token budget across queries to maximize expected accuracy. On the GSM8K arithmetic benchmark, predictive scheduling yields up to 7.9 percentage points of absolute accuracy gain over uniform budgeting at identical token cost, closing over 50\% of the gap to an oracle with perfect foresight. A systematic layer-wise study reveals that middle layers (12 - 17) of the transformer carry the richest signals for size estimation. These results demonstrate that pre-run budget prediction enables fine-grained control of the compute-accuracy trade-off, offering a concrete path toward latency-sensitive, cost-efficient LLM deployments. --- ### Lotus: Efficient LLM Training by Randomized Low-Rank Gradient Projection with Adaptive Subspace Switching Authors: Tianhao Miao, Zhongyuan Bao, Lejun Zhang Published: 2026-02-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.01233v1 Training efficiency in large-scale models is typically assessed through memory consumption, training time, and model performance. Current methods often exhibit trade-offs among these metrics, as optimizing one generally degrades at least one of the others. Addressing this trade-off remains a central challenge in algorithm design. While GaLore enables memory-efficient training by updating gradients in a low-rank subspace, it incurs a comparable extra training time cost due to the Singular Value Decomposition(SVD) process on gradients. In this paper, we propose Lotus, a method that resolves this trade-off by simply modifying the projection process. We propose a criterion that quantifies the displacement of the unit gradient to enable efficient transitions between low-rank gradient subspaces. Experimental results indicate that Lotus is the most efficient method, achieving a 30% reduction in training time and a 40% decrease in memory consumption for gradient and optimizer states. Additionally, it outperforms the baseline method in both pre-training and fine-tuning tasks. --- ### Supervised Fine-Tuning Needs to Unlock the Potential of Token Priority Authors: Zhanming Shen, Zeyu Qin, Jiaqi Hu, Wentao Ye, Hao Chen, Xiaomeng Hu, Haokai Xu, Gang Chen, Yi R. Fung, Haobo Wang Published: 2026-02-01 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.01227v2 The transition from fitting empirical data to achieving true human utility is fundamentally constrained by a granularity mismatch, where fine-grained autoregressive generation is often supervised by coarse or uniform signals. This position paper advocates Token Priority as the essential bridge, formalizing Supervised Fine-Tuning (SFT) not as simple optimization but as a precise distribution reshaping process that aligns raw data with the ideal alignment manifold. We analyze recent breakthroughs through this unified lens, categorizing them into two distinct regimes: Positive Priority for noise filtration and Signed Priority for toxic modes unlearning. We revisit existing progress and limitations, identify key challenges, and suggest directions for future research. --- ### FutureMind: Equipping Small Language Models with Strategic Thinking-Pattern Priors via Adaptive Knowledge Distillation Authors: Shaoxiong Yang, Junting Li, Mengyuan Zhang, Chao Li, Wei Liu, Jian Luan Published: 2026-02-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01222v1 Small Language Models (SLMs) are attractive for cost-sensitive and resource-limited settings due to their efficient, low-latency inference. However, they often struggle with complex, knowledge-intensive tasks that require structured reasoning and effective retrieval. To address these limitations, we propose FutureMind, a modular reasoning framework that equips SLMs with strategic thinking-pattern priors via adaptive knowledge distillation from large language models (LLMs). FutureMind introduces a dynamic reasoning pipeline composed of four key modules: Problem Analysis, Logical Reasoning, Strategy Planning, and Retrieval Guidance. This pipeline is augmented by three distinct retrieval paradigms that decompose complex queries into tractable subproblems, ensuring efficient and accurate retrieval execution. Extensive experiments on multi-hop QA benchmarks, including 2WikiMultihopQA, MuSiQue, Bamboogle, and Frames, demonstrate the superiority of FutureMind. It consistently outperforms strong baselines such as Search-o1, achieving state-of-the-art results under free training conditions across diverse SLM architectures and scales. Beyond empirical gains, our analysis reveals that the process of thinking-pattern distillation is restricted by the cognitive bias bottleneck between the teacher (LLMs) and student (SLMs) models. This provides new perspectives on the transferability of reasoning skills, paving the way for the development of SLMs that combine efficiency with genuine cognitive capability. --- ### Mixture-of-Top-k Attention: Efficient Attention via Scalable Fast Weights Authors: Qishuai Wen, Zhiyuan Huang, Xianghan Meng, Wei He, Chun-Guang Li Published: 2026-02-01 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2602.01219v6 The vanilla self-attention mechanism in Transformers can be viewed as a two-layer fast-weight MLP, whose weights are dynamically induced by inputs and whose hidden dimension is equal to the sequence length $N$. As the context extends, the expressive capacity of such an $N$-width MLP increases, but it becomes unscalable for extremely long sequences. Recently, this fast-weight perspective has motivated the Mixture-of-Experts (MoE) attention mechanism, which partitions the sequence into rigid blocks, treats them as fast-weight experts, and sparsely routes the tokens to them. In this paper, we elevate this perspective to a unifying framework for efficient attention mechanisms, interpreting them as making fast weights scalable through either routing or compression, and organizing them into a five-dimensional taxonomy. Then, we propose Mixture-of-Top-$k$ Attention (MiTA), which employs a small set of landmark queries to gather top-$k$ attended key-value pairs as query-aware and deformable routed experts, while compressing the $N$-width MLP into a narrower shared expert. Consequently, our MiTA improves the flexibility of prior MoE attention from rigid to deformable fast-weight experts, as well as the scalability of prior top-$k$ attention from query-specific set to reusable top-$k$ set. We conduct extensive experiments on vision tasks showing the superior effectiveness and efficiency of our MiTA, and also uncovering intriguing properties such as an emergent token-pruning effect and easy generalization from standard attention. Code is available at https://github.com/QishuaiWen/MiTA. --- ### Mixture-of-Top-k Attention: Efficient Attention via Scalable Fast Weights Authors: Qishuai Wen, Zhiyuan Huang, Xianghan Meng, Wei He, Chun-Guang Li Published: 2026-02-01 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2602.01219v5 The vanilla self-attention mechanism in Transformers can be viewed as a two-layer fast-weight MLP, whose weights are dynamically induced by inputs and whose hidden dimension is equal to the sequence length $N$. As the context extends, the expressive capacity of such an $N$-width MLP increases, but it becomes unscalable for extremely long sequences. Recently, this fast-weight perspective has motivated the Mixture-of-Experts (MoE) attention mechanism, which partitions the sequence into rigid blocks, treats them as fast-weight experts, and sparsely routes the tokens to them. In this paper, we elevate this perspective to a unifying framework for efficient attention mechanisms, interpreting them as making fast weights scalable through either routing or compression, and organizing them into a five-dimensional taxonomy. Then, we propose Mixture-of-Top-$k$ Attention (MiTA), which employs a small set of landmark queries to gather top-$k$ attended key-value pairs as query-aware and deformable routed experts, while compressing the $N$-width MLP into a narrower shared expert. Consequently, our MiTA improves the flexibility of prior MoE attention from rigid to deformable fast-weight experts, as well as the scalability of prior top-$k$ attention from query-specific set to reusable top-$k$ set. We conduct extensive experiments on vision tasks showing the superior effectiveness and efficiency of our MiTA, and also uncovering intriguing properties such as an emergent token-pruning effect and easy generalization from standard attention. Code is available at https://github.com/QishuaiWen/MiTA. --- ### Learning from Anonymized and Incomplete Tabular Data Authors: Lucas Lange, Adrian Böttinger, Victor Christen, Anushka Vidanage, Peter Christen, Erhard Rahm Published: 2026-02-01 Categories: cs.LG, cs.AI, cs.CR, cs.DB arXiv: https://arxiv.org/abs/2602.01217v1 User-driven privacy allows individuals to control whether and at what granularity their data is shared, leading to datasets that mix original, generalized, and missing values within the same records and attributes. While such representations are intuitive for privacy, they pose challenges for machine learning, which typically treats non-original values as new categories or as missing, thereby discarding generalization semantics. For learning from such tabular data, we propose novel data transformation strategies that account for heterogeneous anonymization and evaluate them alongside standard imputation and LLM-based approaches. We employ multiple datasets, privacy configurations, and deployment scenarios, demonstrating that our method reliably regains utility. Our results show that generalized values are preferable to pure suppression, that the best data preparation strategy depends on the scenario, and that consistent data representations are crucial for maintaining downstream utility. Overall, our findings highlight that effective learning is tied to the appropriate handling of anonymized values. --- ### AI Meets Plasticity: A Comprehensive Survey Authors: Hadi Bakhshan, Sima Farshbaf, Junior Ramirez Machado, Fernando Rastellini Canela, Josep Maria Carbonell Published: 2026-02-01 Categories: cond-mat.mtrl-sci, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.01215v1 Artificial intelligence (AI) is rapidly emerging as a new paradigm of scientific discovery, namely data-driven science, across nearly all scientific disciplines. In materials science and engineering, AI has already begun to exert a transformative influence, making it both timely and necessary to examine its interaction with materials plasticity. In this study, we present a holistic survey of the convergence between AI and plasticity, highlighting state-of-the-art AI methodologies employed to discover, construct surrogate models for, and emulate the plastic behavior of materials. From a materials science perspective, we examine cause-and-effect relationships governing plastic deformation, including microstructural characterization and macroscopic responses described through plasticity constitutive models. From the perspective of AI methodology, we review a broad spectrum of applied approaches, ranging from frequentist techniques such as classical machine learning (ML), deep learning (DL), and physics-informed models to probabilistic frameworks that incorporate uncertainty quantification and generative AI methods. These data-driven approaches are discussed in the context of materials characterization and plasticity-related applications. The primary objective of this survey is to develop a comprehensive and well-organized taxonomy grounded in AI methodologies, with particular emphasis on distinguishing critical aspects of these techniques, including model architectures, data requirements, and predictive performance within the specific domain of materials plasticity. By doing so, this work aims to provide a clear road map for researchers and practitioners in the materials community, while offering deeper physical insight and intuition into the role of AI in advancing materials plasticity and characterization, an area of growing importance in the emerging AI-driven era. --- ### SimpleGPT: Improving GPT via A Simple Normalization Strategy Authors: Marco Chen, Xianbiao Qi, Yelin He, Jiaquan Ye, Rong Xiao Published: 2026-02-01 Categories: cs.LG, cs.CL, cs.CV arXiv: https://arxiv.org/abs/2602.01212v1 In this work, we revisit Transformer optimization through the lens of second-order geometry and establish a direct connection between architectural design, activation scale, the Hessian matrix, and the maximum tolerable learning rate. We introduce a simple normalization strategy, termed SimpleNorm, which stabilizes intermediate activation scales by construction. Then, by analyzing the Hessian of the loss with respect to network activations, we theoretically show that SimpleNorm significantly reduces the spectral norm of the Hessian, thereby permitting larger stable learning rates. We validate our theoretical findings through extensive experiments on large GPT models at parameter scales 1B, 1.4B, 7B and 8B. Empirically, SimpleGPT, our SimpleNorm-based network, tolerates learning rates 3$\times$-10$\times$ larger than standard convention, consistently demonstrates strong optimization stability, and achieves substantially better performance than well-established baselines. Specifically, when training 7B-scale models for 60K steps, SimpleGPT achieves a training loss that is 0.08 lower than that of LLaMA2 with QKNorm, reducing the loss from 2.290 to 2.208. Our source code will be released at https://github.com/Ocram7/SimpleGPT. --- ### Chronos: Learning Temporal Dynamics of Reasoning Chains for Test-Time Scaling Authors: Kai Zhang, Jiayi Liao, Chengpeng Li, Ziyuan Xie, Sihang Li, Xiang Wang Published: 2026-02-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.01208v1 Test-Time Scaling (TTS) has emerged as an effective paradigm for improving the reasoning performance of large language models (LLMs). However, existing methods -- most notably majority voting and heuristic token-level scoring -- treat reasoning traces or tokens equally, thereby being susceptible to substantial variations in trajectory quality and localized logical failures. In this work, we introduce \textbf{Chronos}, a lightweight and plug-and-play chronological reasoning scorer that models each trajectory as a time series. Specifically, Chronos learns to capture trajectory features of token probabilities, assigns quality scores accordingly, and employs a weighted voting mechanism. Extensive evaluations on both in-domain and out-of-domain benchmarks demonstrate that Chronos consistently delivers substantial gains across a variety of models, with negligible computational overhead. Notably, Chronos@128 achieves relative improvements of 34.21\% over Pass@1 and 22.70\% over Maj@128 on HMMT25 using Qwen3-4B-Thinking-2507, highlighting its effectiveness. --- ### Not All Preferences Deserve Gradients: Understanding Gradient Utility in Offline Reasoning Alignment Authors: Hui Wu, Hengyi Cai, Jinman Zhao, Xinran Chen, Ziheng Li, Zhejun Zhao, Shuaiqiang Wang, Yuchen Li, Dawei Yin Published: 2026-02-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01207v2 Offline preference optimization aligns reasoning models from fixed chosen--rejected pairs, yet standard methods apply gradient updates from every pair regardless of its training value under the current policy. We argue that this uniform treatment is wasteful and potentially harmful. From the perspective of gradient utility, we show that a pair's contribution depends jointly on informativeness and stability. Pair utility drifts as the policy evolves, high-gradient samples can coincide with high-curvature regions, leading to noisy and destabilizing updates, and the most effective supervision comes from stable confident errors where the model is reliably wrong yet curvature remains low. These findings motivate SAGE (Stability-Aware Gradient Efficiency), which maintains difficulty-stratified candidate pools refreshed during training and selects pairs within each pool by a forward-pass signal-to-curvature score. Only pairs with high current utility receive gradient computation; the rest are excluded from backpropagation. On mathematical reasoning benchmarks across multiple model scales, SAGE outperforms full-data and size-matched baselines while producing substantially smoother optimization trajectories. --- ### Not All Preferences Are Created Equal: Stability-Aware and Gradient-Efficient Alignment for Reasoning Models Authors: Hui Wu, Hengyi Cai, Jinman Zhao, Xinran Chen, Ziheng Li, Zhejun Zhao, Shuaiqiang Wang, Yuchen Li, Dawei Yin Published: 2026-02-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01207v1 Preference-based alignment is pivotal for training large reasoning models; however, standard methods like Direct Preference Optimization (DPO) typically treat all preference pairs uniformly, overlooking the evolving utility of training instances. This static approach often leads to inefficient or unstable optimization, as it wastes computation on trivial pairs with negligible gradients and suffers from noise induced by samples near uncertain decision boundaries. Facing these challenges, we propose SAGE (Stability-Aware Gradient Efficiency), a dynamic framework designed to enhance alignment reliability by maximizing the Signal-to-Noise Ratio of policy updates. Concretely, SAGE integrates a coarse-grained curriculum mechanism that refreshes candidate pools based on model competence with a fine-grained, stability-aware scoring function that prioritizes informative, confident errors while filtering out unstable samples. Experiments on multiple mathematical reasoning benchmarks demonstrate that SAGE significantly accelerates convergence and outperforms static baselines, highlighting the critical role of policy-aware, stability-conscious data selection in reasoning alignment. --- ### Addressing Explainability of Generative AI using SMILE (Statistical Model-agnostic Interpretability with Local Explanations) Authors: Zeinab Dehghani Published: 2026-02-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01206v1 The rapid advancement of generative artificial intelligence has enabled models capable of producing complex textual and visual outputs; however, their decision-making processes remain largely opaque, limiting trust and accountability in high-stakes applications. This thesis introduces gSMILE, a unified framework for the explainability of generative models, extending the Statistical Model-agnostic Interpretability with Local Explanations (SMILE) method to generative settings. gSMILE employs controlled perturbations of textual input, Wasserstein distance metrics, and weighted surrogate modelling to quantify and visualise how specific components of a prompt or instruction influence model outputs. Applied to Large Language Models (LLMs), gSMILE provides fine-grained token-level attribution and generates intuitive heatmaps that highlight influential tokens and reasoning pathways. In instruction-based image editing models, the exact text-perturbation mechanism is employed, allowing for the analysis of how modifications to an editing instruction impact the resulting image. Combined with a scenario-based evaluation strategy grounded in the Operational Design Domain (ODD) framework, gSMILE allows systematic assessment of model behaviour across diverse semantic and environmental conditions. To evaluate explanation quality, we define rigorous attribution metrics, including stability, fidelity, accuracy, consistency, and faithfulness, and apply them across multiple generative architectures. Extensive experiments demonstrate that gSMILE produces robust, human-aligned attributions and generalises effectively across state-of-the-art generative models. These findings highlight the potential of gSMILE to advance transparent, reliable, and responsible deployment of generative AI technologies. --- ### ASTER: Agentic Scaling with Tool-integrated Extended Reasoning Authors: Xuqin Zhang, Quan He, Zhenrui Zheng, Zongzhang Zhang, Xu He, Dong Li Published: 2026-02-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.01204v1 Reinforcement learning (RL) has emerged as a dominant paradigm for eliciting long-horizon reasoning in Large Language Models (LLMs). However, scaling Tool-Integrated Reasoning (TIR) via RL remains challenging due to interaction collapse: a pathological state where models fail to sustain multi-turn tool usage, instead degenerating into heavy internal reasoning with only trivial, post-hoc code verification. We systematically study three questions: (i) how cold-start SFT induces an agentic, tool-using behavioral prior, (ii) how the interaction density of cold-start trajectories shapes exploration and downstream RL outcomes, and (iii) how the RL interaction budget affects learning dynamics and generalization under varying inference-time budgets. We then introduce ASTER (Agentic Scaling with Tool-integrated Extended Reasoning), a framework that circumvents this collapse through a targeted cold-start strategy prioritizing interaction-dense trajectories. We find that a small expert cold-start set of just 4K interaction-dense trajectories yields the strongest downstream performance, establishing a robust prior that enables superior exploration during extended RL training. Extensive evaluations demonstrate that ASTER-4B achieves state-of-the-art results on competitive mathematical benchmarks, reaching 90.0% on AIME 2025, surpassing leading frontier open-source models, including DeepSeek-V3.2-Exp. --- ### Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse Authors: Zizhuo Fu, Wenxuan Zeng, Runsheng Wang, Meng Li Published: 2026-02-01 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.01203v3 Large Language Models (LLMs) often assign disproportionate attention to the first token, a phenomenon known as the attention sink. Several recent approaches aim to address this issue, including Sink Attention in GPT-OSS and Gated Attention in Qwen3-Next. However, a comprehensive analysis of the relationship among these attention mechanisms is lacking. In this work, we provide both theoretical and empirical evidence demonstrating that the sink in Vanilla Attention and Sink Attention naturally construct a Mixture-of-Experts (MoE) mechanism within attention layers. This insight explains the head collapse phenomenon observed in prior work, where only a fixed subset of attention heads contributes to generation. To mitigate head collapse, we propose a sink-aware training algorithm with an auxiliary load balancing loss designed for attention layers. Extensive experiments show that our method achieves effective head load balancing and improves model performance across Vanilla Attention, Sink Attention, and Gated Attention. We hope this study offers a new perspective on attention mechanisms and encourages further exploration of the inherent MoE structure within attention layers. --- ### Workflow-R1: Group Sub-sequence Policy Optimization for Multi-turn Workflow Construction Authors: Mingze Kong, Zikun Qu, Zhongquan Zhou, Pengyu Liang, Xiang Li, Zhiwei Shang, Zhi Hong, Kaiyu Huang, Zhiyong Wang, Zhongxiang Dai Published: 2026-02-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01202v1 The rapid evolution of agentic workflows has demonstrated strong performance of LLM-based agents in addressing complex reasoning tasks. However, existing workflow optimization methods typically formulate workflow synthesis as a static, one-shot code-centric generation problem. This paradigm imposes excessive constraints on the model's coding capabilities and restricts the flexibility required for dynamic problem-solving. In this paper, we present Workflow-R1, a framework that reformulates workflow construction as a multi-turn, natural language-based sequential decision-making process. To resolve the optimization granularity mismatch inherent in such multi-turn interactions, we introduce Group Sub-sequence Policy Optimization (GSsPO). While explicitly tailored to align with the interleaved Think-Action dynamics of agentic reasoning, GSsPO fundamentally functions as a structure-aware RL algorithm generalizable to a broad class of multi-turn agentic sequential decision-making tasks. By recalibrating the optimization unit to the composite sub-sequence, specifically the atomic Think-Action cycle, it aligns gradient updates with the semantic boundaries of these interactions, ensuring robust learning in complex multi-turn reasoning tasks. Through extensive experiments on multiple QA benchmarks, Workflow-R1 outperforms competitive baselines, validating GSsPO as a generalized solution for sequential reasoning and establishing Workflow-R1 as a promising new paradigm for automated workflow optimization. --- ### Med3D-R1: Incentivizing Clinical Reasoning in 3D Medical Vision-Language Models for Abnormality Diagnosis Authors: Haoran Lai, Zihang Jiang, Kun Zhang, Qingsong Yao, Rongsheng Wang, Zhiyang He, Xiaodong Tao, Wei Wei, Shaohua Kevin Zhou Published: 2026-02-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01200v1 Developing 3D vision-language models with robust clinical reasoning remains a challenge due to the inherent complexity of volumetric medical imaging, the tendency of models to overfit superficial report patterns, and the lack of interpretability-aware reward designs. In this paper, we propose Med3D-R1, a reinforcement learning framework with a two-stage training process: Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL). During SFT stage, we introduce a residual alignment mechanism to bridge the gap between high-dimensional 3D features and textual embeddings, and an abnormality re-weighting strategy to emphasize clinically informative tokens and reduce structural bias in reports. In RL stage, we redesign the consistency reward to explicitly promote coherent, step-by-step diagnostic reasoning. We evaluate our method on medical multiple-choice visual question answering using two 3D diagnostic benchmarks, CT-RATE and RAD-ChestCT, where our model attains state-of-the-art accuracies of 41.92\% on CT-RATE and 44.99\% on RAD-ChestCT. These results indicate improved abnormality diagnosis and clinical reasoning and outperform prior methods on both benchmarks. Overall, our approach holds promise for enhancing real-world diagnostic workflows by enabling more reliable and transparent 3D medical vision-language systems. --- ### A State-Transition Framework for Efficient LLM Reasoning Authors: Liang Zhang, Yu Zhao, Longyue Wang, Tianqi Shi, Weihua Luo, Kaifu Zhang, Jinsong Su Published: 2026-02-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01198v1 While Long Chain-of-Thought (CoT) reasoning significantly improves Large Language Models (LLMs) performance on complex reasoning tasks, the substantial computational and memory costs of generating long CoT sequences limit their efficiency and practicality. Existing studies usually enhance the reasoning efficiency of LLMs by compressing CoT sequences. However, this approach conflicts with test-time scaling, limiting the reasoning capacity of LLMs. In this paper, we propose an efficient reasoning framework that models the reasoning process of LLMs as a state-transition process. Specifically, we first apply a linear attention mechanism to estimate the LLM's reasoning state, which records the historical reasoning information from previous reasoning steps. Then, based on the query prompt and the reasoning state, the LLM can efficiently perform the current reasoning step and update the state. With the linear attention, each token in the current reasoning step can directly retrieve relevant historical reasoning information from the reasoning state, without explicitly attending to tokens in previous reasoning steps. In this way, the computational complexity of attention is reduced from quadratic to linear, significantly improving the reasoning efficiency of LLMs. In addition, we propose a state-based reasoning strategy to mitigate the over-thinking issue caused by noisy reasoning steps. Extensive experiments across multiple datasets and model sizes demonstrate that our framework not only improves the reasoning efficiency of LLMs but also enhances their reasoning performance. --- ### Unraveling the Hidden Dynamical Structure in Recurrent Neural Policies Authors: Jin Li, Yue Wu, Mengsha Huang, Yuhao Sun, Hao He, Xianyuan Zhan Published: 2026-02-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01196v2 Recurrent neural policies are widely used in partially observable control and meta-RL tasks. Their abilities to maintain internal memory and adapt quickly to unseen scenarios have offered them unparalleled performance when compared to non-recurrent counterparts. However, until today, the underlying mechanisms for their superior generalization and robustness performance remain poorly understood. In this study, by analyzing the hidden state domain of recurrent policies learned over a diverse set of training methods, model architectures, and tasks, we find that stable cyclic structures consistently emerge during interaction with the environment. Such cyclic structures share a remarkable similarity with \textit{limit cycles} in dynamical system analysis, if we consider the policy and the environment as a joint hybrid dynamical system. Moreover, we uncover that the geometry of such limit cycles also has a structured correspondence with the policies' behaviors. These findings offer new perspectives to explain many nice properties of recurrent policies: the emergence of limit cycles stabilizes both the policies' internal memory and the task-relevant environmental states, while suppressing nuisance variability arising from environmental uncertainty; the geometry of limit cycles also encodes relational structures of behaviors, facilitating easier skill adaptation when facing non-stationary environments. --- ### EMFormer: Efficient Multi-Scale Transformer for Accumulative Context Weather Forecasting Authors: Hao Chen, Tao Han, Jie Zhang, Song Guo, Fenghua Ling, Lei Bai Published: 2026-02-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01194v2 Long-term weather forecasting is critical for socioeconomic planning and disaster preparedness. While recent approaches employ finetuning to extend prediction horizons, they remain constrained by the issues of catastrophic forgetting, error accumulation, and high training overhead. To address these limitations, we present a novel pipeline across pretraining, finetuning and forecasting to enhance long-context modeling while reducing computational overhead. First, we introduce an Efficient Multi-scale Transformer (EMFormer) to extract multi-scale features through a single convolution in both training and inference. Based on the new architecture, we further employ an accumulative context finetuning to improve temporal consistency without degrading short-term accuracy. Additionally, we propose a composite loss that dynamically balances different terms via a sinusoidal weighting, thereby adaptively guiding the optimization trajectory throughout pretraining and finetuning. Experiments show that our approach achieves strong performance in weather forecasting and extreme event prediction, substantially improving long-term forecast accuracy. Moreover, EMFormer demonstrates strong generalization on vision benchmarks (ImageNet-1K and ADE20K) while delivering a 5.69x speedup over conventional multi-scale modules. Code: https://github.com/chenhao-zju/emformer --- ### Bridging Lexical Ambiguity and Vision: A Mini Review on Visual Word Sense Disambiguation Authors: Shashini Nilukshi, Deshan Sumanathilaka Published: 2026-02-01 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2602.01193v1 This paper offers a mini review of Visual Word Sense Disambiguation (VWSD), which is a multimodal extension of traditional Word Sense Disambiguation (WSD). VWSD helps tackle lexical ambiguity in vision-language tasks. While conventional WSD depends only on text and lexical resources, VWSD uses visual cues to find the right meaning of ambiguous words with minimal text input. The review looks at developments from early multimodal fusion methods to new frameworks that use contrastive models like CLIP, diffusion-based text-to-image generation, and large language model (LLM) support. Studies from 2016 to 2025 are examined to show the growth of VWSD through feature-based, graph-based, and contrastive embedding techniques. It focuses on prompt engineering, fine-tuning, and adapting to multiple languages. Quantitative results show that CLIP-based fine-tuned models and LLM-enhanced VWSD systems consistently perform better than zero-shot baselines, achieving gains of up to 6-8\% in Mean Reciprocal Rank (MRR). However, challenges still exist, such as limitations in context, model bias toward common meanings, a lack of multilingual datasets, and the need for better evaluation frameworks. The analysis highlights the growing overlap of CLIP alignment, diffusion generation, and LLM reasoning as the future path for strong, context-aware, and multilingual disambiguation systems. --- ### Autoregressive, Yet Revisable: In Decoding Revision for Secure Code Generation Authors: Chengran Yang, Zichao Wei, Heminghao Deng, Jinfeng Jiang, Zhensu Sun, Ting Zhang, Tianyi Wu, Ming Wen, David Lo Published: 2026-02-01 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2602.01187v2 Large Language Model (LLM) based code generation is predominantly formulated as a strictly monotonic process, appending tokens linearly to an immutable prefix. This formulation contrasts to the cognitive process of programming, which is inherently interleaved with forward generation and on-the-fly revision. While prior works attempt to introduce revision via post-hoc agents or external static tools, they either suffer from high latency or fail to leverage the model's intrinsic semantic reasoning. In this paper, we propose Stream of Revision, a paradigm shift that elevates code generation from a monotonic stream to a dynamic, self-correcting trajectory by leveraging model's intrinsic capabilities. We introduce specific action tokens that enable the model to seamlessly backtrack and edit its own history within a single forward pass. By internalizing the revision loop, our framework Stream of Revision allows the model to activate its latent capabilities just-in-time without external dependencies. Empirical results on secure code generation show that Stream of Revision significantly reduces vulnerabilities with minimal inference overhead. --- ### The Gaussian-Head OFL Family: One-Shot Federated Learning from Client Global Statistics Authors: Fabio Turazza, Marco Picone, Marco Mamei Published: 2026-02-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.01186v2 Classical Federated Learning relies on a multi-round iterative process of model exchange and aggregation between server and clients, with high communication costs and privacy risks from repeated model transmissions. In contrast, one-shot federated learning (OFL) alleviates these limitations by reducing communication to a single round, thereby lowering overhead and enhancing practical deployability. Nevertheless, most existing one-shot approaches remain either impractical or constrained, for example, they often depend on the availability of a public dataset, assume homogeneous client models, or require uploading additional data or model information. To overcome these issues, we introduce the Gaussian-Head OFL (GH-OFL) family, a suite of one-shot federated methods that assume class-conditional Gaussianity of pretrained embeddings. Clients transmit only sufficient statistics (per-class counts and first/second-order moments) and the server builds heads via three components: (i) Closed-form Gaussian heads (NB/LDA/QDA) computed directly from the received statistics; (ii) FisherMix, a linear head with cosine margin trained on synthetic samples drawn in an estimated Fisher subspace; and (iii) Proto-Hyper, a lightweight low-rank residual head that refines Gaussian logits via knowledge distillation on those synthetic samples. In our experiments, GH-OFL methods deliver state-of-the-art robustness and accuracy under strong non-IID skew while remaining strictly data-free. --- ### FedBGS: A Blockchain Approach to Segment Gossip Learning in Decentralized Systems Authors: Fabio Turazza, Marcello Pietri, Marco Picone, Marco Mamei Published: 2026-02-01 Categories: cs.CR, cs.AI, cs.DC, cs.LG arXiv: https://arxiv.org/abs/2602.01185v1 Privacy-Preserving Federated Learning (PPFL) is a Decentralized machine learning paradigm that enables multiple participants to collaboratively train a global model without sharing their data with the integration of cryptographic and privacy-based techniques to enhance the security of the global system. This privacy-oriented approach makes PPFL a highly suitable solution for training shared models in sectors where data privacy is a critical concern. In traditional FL, local models are trained on edge devices, and only model updates are shared with a central server, which aggregates them to improve the global model. However, despite the presence of the aforementioned privacy techniques, in the classical Federated structure, the issue of the server as a single-point-of-failure remains, leading to limitations both in terms of security and scalability. This paper introduces FedBGS, a fully Decentralized Blockchain-based framework that leverages Segmented Gossip Learning through Federated Analytics. The proposed system aims to optimize blockchain usage while providing comprehensive protection against all types of attacks, ensuring both privacy, security and non-IID data handling in Federated environments. --- ### Refining Context-Entangled Content Segmentation via Curriculum Selection and Anti-Curriculum Promotion Authors: Chunming He, Rihan Zhang, Fengyang Xiao, Dingming Zhang, Zhiwen Cao, Sina Farsiu Published: 2026-02-01 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2602.01183v2 Biological learning proceeds from easy to difficult tasks, gradually reinforcing perception and robustness. Inspired by this principle, we address Context-Entangled Content Segmentation (CECS), a challenging setting where objects share intrinsic visual patterns with their surroundings, as in camouflaged object detection. Conventional segmentation networks predominantly rely on architectural enhancements but often ignore the learning dynamics that govern robustness under entangled data distributions. We introduce CurriSeg, a dual-phase learning framework that unifies curriculum and anti-curriculum principles to improve representation reliability. In the Curriculum Selection phase, CurriSeg dynamically selects training data based on the temporal statistics of sample losses, distinguishing hard-but-informative samples from noisy or ambiguous ones, thus enabling stable capability enhancement. In the Anti-Curriculum Promotion phase, we design Spectral-Blindness Fine-Tuning, which suppresses high-frequency components to enforce dependence on low-frequency structural and contextual cues and thus strengthens generalization. Extensive experiments demonstrate that CurriSeg achieves consistent improvements across diverse CECS benchmarks without adding parameters or increasing total training time, offering a principled view of how progression and challenge interplay to foster robust and context-aware segmentation. Code will be released. --- ### Analyzing and Improving Diffusion Models for Time-Series Data Imputation: A Proximal Recursion Perspective Authors: Zhichao Chen, Hao Wang, Fangyikang Wang, Licheng Pan, Zhengnan Li, Yunfei Teng, Haoxuan Li, Zhouchen Lin Published: 2026-02-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01182v1 Diffusion models (DMs) have shown promise for Time-Series Data Imputation (TSDI); however, their performance remains inconsistent in complex scenarios. We attribute this to two primary obstacles: (1) non-stationary temporal dynamics, which can bias the inference trajectory and lead to outlier-sensitive imputations; and (2) objective inconsistency, since imputation favors accurate pointwise recovery whereas DMs are inherently trained to generate diverse samples. To better understand these issues, we analyze DM-based TSDI process through a proximal-operator perspective and uncover that an implicit Wasserstein distance regularization inherent in the process hinders the model's ability to counteract non-stationarity and dissipative regularizer, thereby amplifying diversity at the expense of fidelity. Building on this insight, we propose a novel framework called SPIRIT (Semi-Proximal Transport Regularized time-series Imputation). Specifically, we introduce entropy-induced Bregman divergence to relax the mass preserving constraint in the Wasserstein distance, formulate the semi-proximal transport (SPT) discrepancy, and theoretically prove the robustness of SPT against non-stationarity. Subsequently, we remove the dissipative structure and derive the complete SPIRIT workflow, with SPT serving as the proximal operator. Extensive experiments demonstrate the effectiveness of the proposed SPIRIT approach. --- ### Rethinking the Flow-Based Gradual Domain Adaptation: A Semi-Dual Optimal Transport Perspective Authors: Zhichao Chen, Zhan Zhuang, Yunfei Teng, Hao Wang, Fangyikang Wang, Zhengnan Li, Tianqiao Liu, Haoxuan Li, Zhouchen Lin Published: 2026-02-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01179v2 Gradual domain adaptation (GDA) aims to mitigate domain shift by progressively adapting models from the source domain to the target domain via intermediate domains. However, real intermediate domains are often unavailable or ineffective, necessitating the synthesis of intermediate samples. Flow-based models have recently been used for this purpose by interpolating between source and target distributions. Notably, their training typically relies on sample-based log-likelihood estimation, which can discard useful information and thus degrade GDA performance. The key to addressing this limitation is constructing the intermediate domains via samples directly. To this end, we propose an Entropy-regularized Semi-dual Unbalanced Optimal Transport (E-SUOT) framework to construct intermediate domains. Specifically, we reformulate flow-based GDA as a Lagrangian dual problem and derive an equivalent semi-dual objective that circumvents the need for likelihood estimation. However, the dual problem leads to an unstable min-max training procedure. To alleviate this issue, we further introduce the entropy regularization to convert it into a more stable sequential optimization procedure. Based on this, we propose a novel GDA training framework and provide theoretical analysis in terms of stability and generalization. Finally, extensive experiments are conducted to demonstrate the efficacy of the E-SUOT framework. --- ### Privacy Implies Stability: Information-Theoretic Generalization Bounds for Quantum Learning Authors: Ayanava Dasgupta, Naqueeb Ahmad Warsi, Masahito Hayashi Published: 2026-02-01 Categories: quant-ph, cs.IT, cs.LG arXiv: https://arxiv.org/abs/2602.01177v3 We develop an information-theoretic framework connecting stability, privacy, and generalization for quantum learning algorithms. Learning procedures are modeled as quantum instruments with classical-quantum outputs, and losses are represented by observables. We prove that under a classical-quantum sub-Gaussian condition, an information-theoretic stability measure controls the expected generalization error. Furthermore, we establish a high-probability generalization bound using quantum Rényi divergences to manage higher-order dependencies under non-commutativity. In the trusted Data Processor setting, quantum differential privacy (QDP) provides a mechanism for stability. We show that one-neighbor QDP strictly bounds the information leaked by the classical-quantum output. Combining this with our stability theorem yields a direct privacy-to-generalization guarantee. We also explore an untrusted Data Processor setting. Here, output privacy alone is insufficient since an adversarial processor could perform a highly informative procedure before applying noisy post-processing. To combat this, we introduce Information-Theoretic Admissibility (ITA), a certification condition ensuring the prescribed procedure is not just a degraded version of a strictly more informative, physically allowed operation on the encoded ensemble. We prove a fundamental separation: while admissibility and privacy are in strong tension in classical models, quantum non-orthogonality makes them compatible. A quantum measurement can be ITA - exhausting all relevant accessible information - without perfectly recovering the classical dataset. We illustrate this separation through a concrete quantum ITA example. ---