# Research Papers — 2026-02 · bundle page 200 of 210 > Full-text markdown of papers #9951–10000 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/200 [← Previous bundle](https://aipod.dev/papers/archive/2026-02/page/199.md) [Next bundle →](https://aipod.dev/papers/archive/2026-02/page/201.md) ### Autocorrelated Optimize-via-Estimate: Predict-then-Optimize versus Finite-sample Optimal Authors: Zichun Wang, Gar Goei Loke, Ruiting Zuo Published: 2026-02-02 Categories: cs.LG, math.OC arXiv: https://arxiv.org/abs/2602.01877v1 Models that directly optimize for out-of-sample performance in the finite-sample regime have emerged as a promising alternative to traditional estimate-then-optimize approaches in data-driven optimization. In this work, we compare their performance in the context of autocorrelated uncertainties, specifically, under a Vector Autoregressive Moving Average VARMA(p,q) process. We propose an autocorrelated Optimize-via-Estimate (A-OVE) model that obtains an out-of-sample optimal solution as a function of sufficient statistics, and propose a recursive form for computing its sufficient statistics. We evaluate these models on a portfolio optimization problem with trading costs. A-OVE achieves low regret relative to a perfect information oracle, outperforming predict-then-optimize machine learning benchmarks. Notably, machine learning models with higher accuracy can have poorer decision quality, echoing the growing literature in data-driven optimization. Performance is retained under small mis-specification. --- ### PretrainRL: Alleviating Factuality Hallucination of Large Language Models at the Beginning Authors: Langming Liu, Kangtao Lv, Haibin Chen, Weidong Zhang, Yejing Wang, Shilei Liu, Xin Tong, Yujin Yuan, Yongwei Wang, Wenbo Su, Bo Zheng Published: 2026-02-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.01875v1 Large language models (LLMs), despite their powerful capabilities, suffer from factual hallucinations where they generate verifiable falsehoods. We identify a root of this issue: the imbalanced data distribution in the pretraining corpus, which leads to a state of "low-probability truth" and "high-probability falsehood". Recent approaches, such as teaching models to say "I don't know" or post-hoc knowledge editing, either evade the problem or face catastrophic forgetting. To address this issue from its root, we propose \textbf{PretrainRL}, a novel framework that integrates reinforcement learning into the pretraining phase to consolidate factual knowledge. The core principle of PretrainRL is "\textbf{debiasing then learning}." It actively reshapes the model's probability distribution by down-weighting high-probability falsehoods, thereby making "room" for low-probability truths to be learned effectively. To enable this, we design an efficient negative sampling strategy to discover these high-probability falsehoods and introduce novel metrics to evaluate the model's probabilistic state concerning factual knowledge. Extensive experiments on three public benchmarks demonstrate that PretrainRL significantly alleviates factual hallucinations and outperforms state-of-the-art methods. --- ### Grappa: Gradient-Only Communication for Scalable Graph Neural Network Training Authors: Chongyang Xu, Christoph Siebenbrunner, Laurent Bindschaedler Published: 2026-02-02 Categories: cs.DC, cs.LG arXiv: https://arxiv.org/abs/2602.01872v2 Cross-partition edges dominate the cost of distributed GNN training: fetching remote features and activations per iteration overwhelms the network as graphs deepen and partition counts grow. Grappa is a distributed GNN training framework that enforces gradient-only communication: during each iteration, partitions train in isolation and exchange only gradients for the global update. To recover accuracy lost to isolation, Grappa (i) periodically repartitions to expose new neighborhoods and (ii) applies a lightweight coverage-corrected gradient aggregation inspired by importance sampling. We present an asymptotically unbiased estimator for gradient correction, which we use to develop a minimum-distance batch-level variant that is compatible with common deep-learning packages. We also introduce a shrinkage version that improves stability in practice. Empirical results on real and synthetic graphs show that Grappa trains GNNs 4x faster on average (up to 13x) than state-of-the-art systems, achieves better accuracy especially for deeper models, and sustains training at the trillion-edge scale on commodity hardware. Grappa is model-agnostic, supports full-graph and mini-batch training, and does not rely on high-bandwidth interconnects or caching. --- ### Skill-Pro: Learning Reusable Skills from Experience via Non-Parametric PPO for LLM Agents Authors: Qirui Mi, Zhijian Ma, Mengyue Yang, Haoxuan Li, Yisen Wang, Haifeng Zhang, Jun Wang Published: 2026-02-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01869v3 LLM-driven agents excel at sequential decision-making but often rely on on-the-fly reasoning, re-deriving solutions even in recurring scenarios. This insufficient experience reuse leads to computational redundancy and instability. To bridge this gap, we propose Skill-Pro, a framework enabling agents to autonomously learn reusable procedural skills from interaction experiences without parameter updates. By formalizing a Skill-MDP, Skill-Pro transforms passive episodic narratives into executable Skills defined by activation, execution, and termination conditions to ensure executability. To achieve reliable reusability without capability degradation, we introduce Non-Parametric PPO, which leverages semantic gradients for high-quality candidate generation and a PPO Gate for robust Skill verification. Through score-based maintenance, Skill-Pro sustains compact, high-quality procedural memory. Experimental results across in-domain, cross-task, and cross-agent scenarios demonstrate that Skill-Pro achieves superior reuse rates and significant gains with extreme memory compression. Visualized evolutionary trajectories and Skill distributions further reveal how Skill-Pro transparently accumulates, refines, and reuses procedural knowledge to facilitate long-term autonomy. --- ### GRAB: An LLM-Inspired Sequence-First Click-Through Rate Prediction Modeling Paradigm Authors: Shaopeng Chen, Chuyue Xie, Huimin Ren, Shaozong Zhang, Han Zhang, Ruobing Cheng, Zhiqiang Cao, Zehao Ju, Yu Gao, Jie Ding, Xiaodong Chen, Xuewu Jiao et al. Published: 2026-02-02 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2602.01865v2 Traditional Deep Learning Recommendation Models (DLRMs) face increasing bottlenecks in performance and efficiency, often struggling with generalization and long-sequence modeling. Inspired by the scaling success of Large Language Models (LLMs), we propose Generative Ranking for Ads at Baidu (GRAB), an end-to-end generative framework for Click-Through Rate (CTR) prediction. GRAB integrates a novel Causal Action-aware Multi-channel Attention (CamA) mechanism to effectively capture temporal dynamics and specific action signals within user behavior sequences. Full-scale online deployment demonstrates that GRAB significantly outperforms established DLRMs, delivering a 3.05% increase in revenue and a 3.49% rise in CTR. Furthermore, the model demonstrates desirable scaling behavior: its expressive power shows a monotonic and approximately linear improvement as longer interaction sequences are utilized. --- ### Trust but Verify: Adaptive Conditioning for Reference-Based Diffusion Super-Resolution via Implicit Reference Correlation Modeling Authors: Yuan Wang, Yuhao Wan, Siming Zheng, Bo Li, Qibin Hou, Peng-Tao Jiang Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01864v1 Recent works have explored reference-based super-resolution (RefSR) to mitigate hallucinations in diffusion-based image restoration. A key challenge is that real-world degradations make correspondences between low-quality (LQ) inputs and reference (Ref) images unreliable, requiring adaptive control of reference usage. Existing methods either ignore LQ-Ref correlations or rely on brittle explicit matching, leading to over-reliance on misleading references or under-utilization of valuable cues. To address this, we propose Ada-RefSR, a single-step diffusion framework guided by a "Trust but Verify" principle: reference information is leveraged when reliable and suppressed otherwise. Its core component, Adaptive Implicit Correlation Gating (AICG), employs learnable summary tokens to distill dominant reference patterns and capture implicit correlations with LQ features. Integrated into the attention backbone, AICG provides lightweight, adaptive regulation of reference guidance, serving as a built-in safeguard against erroneous fusion. Experiments on multiple datasets demonstrate that Ada-RefSR achieves a strong balance of fidelity, naturalness, and efficiency, while remaining robust under varying reference alignment. --- ### Transformers as Measure-Theoretic Associative Memory: A Statistical Perspective and Minimax Optimality Authors: Ryotaro Kawata, Taiji Suzuki Published: 2026-02-02 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2602.01863v1 Transformers excel through content-addressable retrieval and the ability to exploit contexts of, in principle, unbounded length. We recast associative memory at the level of probability measures, treating a context as a distribution over tokens and viewing attention as an integral operator on measures. Concretely, for mixture contexts $ν= I^{-1} \sum_{i=1}^I μ^{(i^*)}$ and a query $x_{\mathrm{q}}(i^*)$, the task decomposes into (i) recall of the relevant component $μ^{(i^*)}$ and (ii) prediction from $(μ_{i^*},x_\mathrm{q})$. We study learned softmax attention (not a frozen kernel) trained by empirical risk minimization and show that a shallow measure-theoretic Transformer composed with an MLP learns the recall-and-predict map under a spectral assumption on the input densities. We further establish a matching minimax lower bound with the same rate exponent (up to multiplicative constants), proving sharpness of the convergence order. The framework offers a principled recipe for designing and analyzing Transformers that recall from arbitrarily long, distributional contexts with provable generalization guarantees. --- ### RIR-Former: Coordinate-Guided Transformer for Continuous Reconstruction of Room Impulse Responses Authors: Shaoheng Xu, Chunyi Sun, Jihui Zhang, Prasanga N. Samarasinghe, Thushara D. Abhayapala Published: 2026-02-02 Categories: eess.AS, cs.LG arXiv: https://arxiv.org/abs/2602.01861v3 Room impulse responses (RIRs) are essential for many acoustic signal processing tasks, yet measuring them densely across space is often impractical. In this work, we propose RIR-Former, a grid-free, one-step feed-forward model for RIR reconstruction. By introducing a sinusoidal encoding module into a transformer backbone, our method effectively incorporates microphone position information, enabling interpolation at arbitrary array locations. Furthermore, a segmented multi-branch decoder is designed to separately handle early reflections and late reverberation, improving reconstruction across the entire RIR. Experiments on diverse simulated acoustic environments demonstrate that RIR-Former consistently outperforms state-of-the-art baselines in terms of normalized mean square error (NMSE) and cosine distance (CD), under varying missing rates and array configurations. These results highlight the potential of our approach for practical deployment and motivate future work on scaling from randomly spaced linear arrays to complex array geometries, dynamic acoustic scenes, and real-world environments. --- ### SOPRAG: Multi-view Graph Experts Retrieval for Industrial Standard Operating Procedures Authors: Liangtao Lin, Zhaomeng Zhu, Tianwei Zhang, Yonggang Wen Published: 2026-02-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01858v1 Standard Operating Procedures (SOPs) are essential for ensuring operational safety and consistency in industrial environments. However, retrieving and following these procedures presents unique challenges, such as rigid proprietary structures, condition-dependent relevance, and actionable execution requirement, which standard semantic-driven Retrieval-Augmented Generation (RAG) paradigms fail to address. Inspired by the Mixture-of-Experts (MoE) paradigm, we propose SOPRAG, a novel framework specifically designed to address the above pain points in SOP retrieval. SOPRAG replaces flat chunking with specialized Entity, Causal, and Flow graph experts to resolve industrial structural and logical complexities. To optimize and coordinate these experts, we propose a Procedure Card layer that prunes the search space to eliminate computational noise, and an LLM-Guided gating mechanism that dynamically weights these experts to align retrieval with operator intent. To address the scarcity of domain-specific data, we also introduce an automated, multi-agent workflow for benchmark construction. Extensive experiments across four industrial domains demonstrate that SOPRAG significantly outperforms strong lexical, dense, and graph-based RAG baselines in both retrieval accuracy and response utility, achieving perfect execution scores in real-world critical tasks. --- ### Time2Vec Transformer for Robust Gesture Recognition from Low-Density sEMG Authors: Blagoj Hristov, Hristijan Gjoreski, Vesna Ojleska Latkoska, Gorjan Nadzinski Published: 2026-02-02 Categories: cs.LG, cs.AI, eess.SP arXiv: https://arxiv.org/abs/2602.01855v2 Accurate and responsive myoelectric prosthesis control typically relies on complex, dense multi-sensor arrays, which limits consumer accessibility. This paper presents a novel, data-efficient deep learning framework designed to achieve precise and accurate control using minimal sensor hardware. Leveraging an external dataset of 8 subjects, our approach implements a hybrid Transformer optimized for sparse, two-channel surface electromyography (sEMG). Unlike standard architectures that use fixed positional encodings, we integrate Time2Vec learnable temporal embeddings to capture the stochastic temporal warping inherent in biological signals. Furthermore, we employ a normalized additive fusion strategy that aligns the latent distributions of spatial and temporal features, preventing the destructive interference common in standard implementations. A two-stage curriculum learning protocol is utilized to ensure robust feature extraction despite data scarcity. The proposed architecture achieves a state-of-the-art multi-subject F1-score of 95.7% $\pm$ 0.20% for a 10-class movement set, statistically outperforming both a standard Transformer with fixed encodings and a recurrent CNN-LSTM model. Architectural optimization reveals that a balanced allocation of model capacity between spatial and temporal dimensions yields the highest stability. Furthermore, while direct transfer to a new unseen subject led to poor accuracy due to domain shifts, a rapid calibration protocol utilizing only two trials per gesture recovered performance from 21.0% $\pm$ 2.98% to 96.9% $\pm$ 0.52%. By validating that high-fidelity temporal embeddings can compensate for low spatial resolution, this work challenges the necessity of high-density sensing. The proposed framework offers a robust, cost-effective blueprint for next-generation prosthetic interfaces capable of rapid personalization. --- ### Fact or Fake? Assessing the Role of Deepfake Detectors in Multimodal Misinformation Detection Authors: A S M Sharifuzzaman Sagar, Mohammed Bennamoun, Farid Boussaid, Naeha Sharif, Lian Xu, Shaaban Sahmoud, Ali Kishk Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01854v1 In multimodal misinformation, deception usually arises not just from pixel-level manipulations in an image, but from the semantic and contextual claim jointly expressed by the image-text pair. Yet most deepfake detectors, engineered to detect pixel-level forgeries, do not account for claim-level meaning, despite their growing integration in automated fact-checking (AFC) pipelines. This raises a central scientific and practical question: Do pixel-level detectors contribute useful signal for verifying image-text claims, or do they instead introduce misleading authenticity priors that undermine evidence-based reasoning? We provide the first systematic analysis of deepfake detectors in the context of multimodal misinformation detection. Using two complementary benchmarks, MMFakeBench and DGM4, we evaluate: (1) state-of-the-art image-only deepfake detectors, (2) an evidence-driven fact-checking system that performs tool-guided retrieval via Monte Carlo Tree Search (MCTS) and engages in deliberative inference through Multi-Agent Debate (MAD), and (3) a hybrid fact-checking system that injects detector outputs as auxiliary evidence. Results across both benchmark datasets show that deepfake detectors offer limited standalone value, achieving F1 scores in the range of 0.26-0.53 on MMFakeBench and 0.33-0.49 on DGM4, and that incorporating their predictions into fact-checking pipelines consistently reduces performance by 0.04-0.08 F1 due to non-causal authenticity assumptions. In contrast, the evidence-centric fact-checking system achieves the highest performance, reaching F1 scores of approximately 0.81 on MMFakeBench and 0.55 on DGM4. Overall, our findings demonstrate that multimodal claim verification is driven primarily by semantic understanding and external evidence, and that pixel-level artifact signals do not reliably enhance reasoning over real-world image-text misinformation. --- ### Designing Time Series Experiments in A/B Testing with Transformer Reinforcement Learning Authors: Xiangkun Wu, Qianglin Wen, Yingying Zhang, Hongtu Zhu, Ting Li, Chengchun Shi Published: 2026-02-02 Categories: cs.LG, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2602.01853v1 A/B testing has become a gold standard for modern technological companies to conduct policy evaluation. Yet, its application to time series experiments, where policies are sequentially assigned over time, remains challenging. Existing designs suffer from two limitations: (i) they do not fully leverage the entire history for treatment allocation; (ii) they rely on strong assumptions to approximate the objective function (e.g., the mean squared error of the estimated treatment effect) for optimizing the design. We first establish an impossibility theorem showing that failure to condition on the full history leads to suboptimal designs, due to the dynamic dependencies in time series experiments. To address both limitations simultaneously, we next propose a transformer reinforcement learning (RL) approach which leverages transformers to condition allocation on the entire history and employs RL to directly optimize the MSE without relying on restrictive assumptions. Empirical evaluations on synthetic data, a publicly available dispatch simulator, and a real-world ridesharing dataset demonstrate that our proposal consistently outperforms existing designs. --- ### FUPareto: Bridging the Forgetting-Utility Gap in Federated Unlearning via Pareto Augmented Optimization Authors: Zeyan Wang, Zhengmao Liu, Yongxin Cai, Chi Li, Xiaoying Tang, Jingchao Chen, Zibin Pan, Jing Qiu Published: 2026-02-02 Categories: cs.LG, cs.DC arXiv: https://arxiv.org/abs/2602.01852v1 Federated Unlearning (FU) aims to efficiently remove the influence of specific client data from a federated model while preserving utility for the remaining clients. However, three key challenges remain: (1) existing unlearning objectives often compromise model utility or increase vulnerability to Membership Inference Attacks (MIA); (2) there is a persistent conflict between forgetting and utility, where further unlearning inevitably harms retained performance; and (3) support for concurrent multi-client unlearning is poor, as gradient conflicts among clients degrade the quality of forgetting. To address these issues, we propose FUPareto, an efficient unlearning framework via Pareto-augmented optimization. We first introduce the Minimum Boundary Shift (MBS) Loss, which enforces unlearning by suppressing the target class logit below the highest non-target class logit; this can improve the unlearning efficiency and mitigate MIA risks. During the unlearning process, FUPareto performs Pareto improvement steps to preserve model utility and executes Pareto expansion to guarantee forgetting. Specifically, during Pareto expansion, the framework integrates a Null-Space Projected Multiple Gradient Descent Algorithm (MGDA) to decouple gradient conflicts. This enables effective, fair, and concurrent unlearning for multiple clients while minimizing utility degradation. Extensive experiments across diverse scenarios demonstrate that FUPareto consistently outperforms state-of-the-art FU methods in both unlearning efficacy and retained utility. --- ### How Well Do Models Follow Visual Instructions? VIBE: A Systematic Benchmark for Visual Instruction-Driven Image Editing Authors: Huanyu Zhang, Xuehai Bai, Chengzu Li, Chen Liang, Haochen Tian, Haodong Li, Ruichuan An, Yifan Zhang, Anna Korhonen, Zhang Zhang, Liang Wang, Tieniu Tan Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01851v2 Recent generative models have achieved remarkable progress in image editing. However, existing systems and benchmarks remain largely text-guided. In contrast, human communication is inherently multimodal, where visual instructions such as sketches efficiently convey spatial and structural intent. To address this gap, we introduce VIBE, the Visual Instruction Benchmark for Image Editing with a three-level interaction hierarchy that captures deictic grounding, morphological manipulation, and causal reasoning. Across these levels, we curate high-quality and diverse test cases that reflect progressively increasing complexity in visual instruction following. We further propose a robust LMM-as-a-judge evaluation framework with task-specific metrics to enable scalable and fine-grained assessment. Through a comprehensive evaluation of 17 representative open-source and proprietary image editing models, we find that proprietary models exhibit early-stage visual instruction-following capabilities and consistently outperform open-source models. However, performance degrades markedly with increasing task difficulty even for the strongest systems, highlighting promising directions for future research. --- ### WS-IMUBench: Can Weakly Supervised Methods from Audio, Image, and Video Be Adapted for IMU-based Temporal Action Localization? Authors: Pei Li, Jiaxi Yin, Lei Ouyang, Shihan Pan, Ge Wang, Han Ding, Fei Wang Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01850v1 IMU-based Human Activity Recognition (HAR) has enabled a wide range of ubiquitous computing applications, yet its dominant clip classification paradigm cannot capture the rich temporal structure of real-world behaviors. This motivates a shift toward IMU Temporal Action Localization (IMU-TAL), which predicts both action categories and their start/end times in continuous streams. However, current progress is strongly bottlenecked by the need for dense, frame-level boundary annotations, which are costly and difficult to scale. To address this bottleneck, we introduce WS-IMUBench, a systematic benchmark study of weakly supervised IMU-TAL (WS-IMU-TAL) under only sequence-level labels. Rather than proposing a new localization algorithm, we evaluate how well established weakly supervised localization paradigms from audio, image, and video transfer to IMU-TAL under only sequence-level labels. We benchmark seven representative weakly supervised methods on seven public IMU datasets, resulting in over 3,540 model training runs and 7,080 inference evaluations. Guided by three research questions on transferability, effectiveness, and insights, our findings show that (i) transfer is modality-dependent, with temporal-domain methods generally more stable than image-derived proposal-based approaches; (ii) weak supervision can be competitive on favorable datasets (e.g., with longer actions and higher-dimensional sensing); and (iii) dominant failure modes arise from short actions, temporal ambiguity, and proposal quality. Finally, we outline concrete directions for advancing WS-IMU-TAL (e.g., IMU-specific proposal generation, boundary-aware objectives, and stronger temporal reasoning). Beyond individual results, WS-IMUBench establishes a reproducible benchmarking template, datasets, protocols, and analyses, to accelerate community-wide progress toward scalable WS-IMU-TAL. --- ### Self-Rewarding Sequential Monte Carlo for Masked Diffusion Language Models Authors: Ziwei Luo, Ziqi Jin, Lei Wang, Lidong Bing, Thomas B. Schön Published: 2026-02-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01849v1 This work presents self-rewarding sequential Monte Carlo (SMC), an inference-time scaling algorithm enabling effective sampling of masked diffusion language models (MDLMs). Our algorithm stems from the observation that most existing MDLMs rely on a confidence-based sampling strategy, where only tokens with the highest prediction confidence are preserved at each step. This restricts the generation to a noise-sensitive, greedy decoding paradigm, resulting in an inevitable collapse in the diversity of possible paths. We address this problem by launching multiple interacting diffusion processes in parallel, referred to as particles, for trajectory exploration. Importantly, we introduce the trajectory-level confidence as a self-rewarding signal for assigning particle importance weights. During sampling, particles are iteratively weighted and resampled to systematically steer generation towards globally confident, high-quality samples. Our self-rewarding SMC is verified on various masked diffusion language models and benchmarks, achieving significant improvement without extra training or reward guidance, while effectively converting parallel inference capacity into improved sampling quality. Our code is available at https://github.com/Algolzw/self-rewarding-smc. --- ### ROMA: Recursive Open Meta-Agent Framework for Long-Horizon Multi-Agent Systems Authors: Salaheddin Alzu'bi, Baran Nama, Arda Kaz, Anushri Eswaran, Weiyuan Chen, Sarvesh Khetan, Rishab Bala, Tu Vu, Sewoong Oh Published: 2026-02-02 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2602.01848v2 Current agentic frameworks underperform on long-horizon tasks. As reasoning depth increases, sequential orchestration becomes brittle, context windows impose hard limits that degrade performance, and opaque execution traces make failures difficult to localize or debug. We introduce ROMA (Recursive Open Meta-Agents), a domain-agnostic framework that addresses these limitations through recursive task decomposition and structured aggregation. ROMA decomposes goals into dependency-aware subtask trees that can be executed in parallel, while aggregation compresses and validates intermediate results to control context growth. Our framework standardizes agent construction around four modular roles --Atomizer (which decides whether a task should be decomposed), Planner, Executor, and Aggregator -- which cleanly separate orchestration from model selection and enable transparent, hierarchical execution traces. This design supports heterogeneous multi-agent systems that mix models and tools according to cost, latency, and capability. To adapt ROMA to specific tasks without fine-tuning, we further introduce GEPA$+$, an improved Genetic-Pareto prompt proposer that searches over prompts within ROMA's component hierarchy while preserving interface contracts. We show that ROMA, combined with GEPA+, delivers leading system-level performance on reasoning and long-form generation benchmarks. On SEAL-0, which evaluates reasoning over conflicting web evidence, ROMA instantiated with GLM-4.6 improves accuracy by 9.9\% over Kimi-Researcher. On EQ-Bench, a long-form writing benchmark, ROMA enables DeepSeek-V3 to match the performance of leading closed-source models such as Claude Sonnet 4.5. Our results demonstrate that recursive, modular agent architectures can scale reasoning depth while remaining interpretable, flexible, and model-agnostic. --- ### No Generation without Representation: Efficient Causal Protein Language Models Enable Zero-Shot Fitness Estimation Authors: Furkan Eris Published: 2026-02-02 Categories: cs.LG, q-bio.QM arXiv: https://arxiv.org/abs/2602.01845v1 Protein language models (PLMs) face a fundamental divide: masked language models (MLMs) excel at fitness prediction while causal models enable generation, forcing practitioners to maintain separate architectures. We introduce \textbf{Proust}, a 309M-parameter causal PLM that bridges this gap through architectural innovations adapted from recent LLM research, including grouped-query attention with shared K/V projections, cross-layer value residuals, and depthwise causal convolutions. Trained on 33B tokens in 40 B200 GPU-hours, Proust achieves Spearman $ρ= 0.390$ on ProteinGym substitutions, competitive with MLMs requiring 50--200$\times$ the compute. On indels, Proust sets a new state-of-the-art, outperforming models up to 20$\times$ larger. On EVEREST viral fitness benchmarks, it approaches structure-aware methods using sequence alone. These powerful representations position Proust in a sweet spot as it also retains native generative capabilities that MLMs lack by design. Interpretability analysis reveals that per-position entropy variance predicts, to an extent, when retrieval augmentation helps and hurts. Such insights can grow in both quantity and quality at scale and inform capabilities such as test-time scaling. Code and weights are available at https://github.com/Furkan9015/proust-inference --- ### CloDS: Visual-Only Unsupervised Cloth Dynamics Learning in Unknown Conditions Authors: Yuliang Zhan, Jian Li, Wenbing Huang, Wenbing Huang, Yang Liu, Hao Sun Published: 2026-02-02 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.01844v3 Deep learning has demonstrated remarkable capabilities in simulating complex dynamic systems. However, existing methods require known physical properties as supervision or inputs, limiting their applicability under unknown conditions. To explore this challenge, we introduce Cloth Dynamics Grounding (CDG), a novel scenario for unsupervised learning of cloth dynamics from multi-view visual observations. We further propose Cloth Dynamics Splatting (CloDS), an unsupervised dynamic learning framework designed for CDG. CloDS adopts a three-stage pipeline that first performs video-to-geometry grounding and then trains a dynamics model on the grounded meshes. To cope with large non-linear deformations and severe self-occlusions during grounding, we introduce a dual-position opacity modulation that supports bidirectional mapping between 2D observations and 3D geometry via mesh-based Gaussian splatting in video-to-geometry grounding stage. It jointly considers the absolute and relative position of Gaussian components. Comprehensive experimental evaluations demonstrate that CloDS effectively learns cloth dynamics from visual data while maintaining strong generalization capabilities for unseen configurations. Our code is available at https://github.com/whynot-zyl/CloDS. Visualization results are available at https://github.com/whynot-zyl/CloDS_video}.%\footnote{As in this example. --- ### SPIRIT: Adapting Vision Foundation Models for Unified Single- and Multi-Frame Infrared Small Target Detection Authors: Qian Xu, Xi Li, Fei Gao, Jie Guo, Haojuan Yuan, Shuaipeng Fan, Mingjin Zhang Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01843v1 Infrared small target detection (IRSTD) is crucial for surveillance and early-warning, with deployments spanning both single-frame analysis and video-mode tracking. A practical solution should leverage vision foundation models (VFMs) to mitigate infrared data scarcity, while adopting a memory-attention-based temporal propagation framework that unifies single- and multi-frame inference. However, infrared small targets exhibit weak radiometric signals and limited semantic cues, which differ markedly from visible-spectrum imagery. This modality gap makes direct use of semantics-oriented VFMs and appearance-driven cross-frame association unreliable for IRSTD: hierarchical feature aggregation can submerge localized target peaks, and appearance-only memory attention becomes ambiguous, leading to spurious clutter associations. To address these challenges, we propose SPIRIT, a unified and VFM-compatible framework that adapts VFMs to IRSTD via lightweight physics-informed plug-ins. Spatially, PIFR refines features by approximating rank-sparsity decomposition to suppress structured background components and enhance sparse target-like signals. Temporally, PGMA injects history-derived soft spatial priors into memory cross-attention to constrain cross-frame association, enabling robust video detection while naturally reverting to single-frame inference when temporal context is absent. Experiments on multiple IRSTD benchmarks show consistent gains over VFM-based baselines and SOTA performance. --- ### Prism: Efficient Test-Time Scaling via Hierarchical Search and Self-Verification for Discrete Diffusion Language Models Authors: Jinbin Bai, Yixuan Li, Yuchen Zhu, Yi Xin, Qingyu Shi, Aosong Feng, Xiaohong Liu, Molei Tao, Jianru Xue, Xiangtai Li, Ming-Hsuan Yang Published: 2026-02-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01842v3 Inference-time compute has re-emerged as a practical way to improve LLM reasoning. Most test-time scaling (TTS) algorithms rely on autoregressive decoding, which is ill-suited to discrete diffusion language models (dLLMs) due to their parallel decoding over the entire sequence. As a result, developing effective and efficient TTS methods to unlock dLLMs' full generative potential remains an underexplored challenge. To address this, we propose Prism (Pruning, Remasking, and Integrated Self-verification Method), an efficient TTS framework for dLLMs that (i) performs Hierarchical Trajectory Search (HTS) which dynamically prunes and reallocates compute in an early-to-mid denoising window, (ii) introduces Local branching with partial remasking to explore diverse implementations while preserving high-confidence tokens, and (iii) replaces external verifiers with Self-Verified Feedback (SVF) obtained via self-evaluation prompts on intermediate completions. Across four mathematical reasoning and code generation benchmarks on three dLLMs, including LLaDA 8B Instruct, Dream 7B Instruct, and LLaDA 2.0-mini, our Prism achieves a favorable performance-efficiency trade-off, matching best-of-N performance with substantially fewer function evaluations (NFE). The code is released at https://github.com/viiika/Prism. --- ### Read As Human: Compressing Context via Parallelizable Close Reading and Skimming Authors: Jiwei Tang, Shilei Liu, Zhicheng Zhang, Qingsong Lv, Runsong Zhao, Tingwei Lu, Langming Liu, Haibin Chen, Yujin Yuan, Hai-Tao Zheng, Wenbo Su, Bo Zheng Published: 2026-02-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.01840v2 Large Language Models (LLMs) demonstrate exceptional capability across diverse tasks. However, their deployment in long-context scenarios is hindered by two challenges: computational inefficiency and redundant information. We propose RAM (Read As HuMan), a context compression framework that adopts an adaptive hybrid reading strategy, to address these challenges. Inspired by human reading behavior (i.e., close reading important content while skimming less relevant content), RAM partitions the context into segments and encodes them with the input query in parallel. High-relevance segments are fully retained (close reading), while low-relevance ones are query-guided compressed into compact summary vectors (skimming). Both explicit textual segments and implicit summary vectors are concatenated and fed into decoder to achieve both superior performance and natural language format interpretability. To refine the decision boundary between close reading and skimming, we further introduce a contrastive learning objective based on positive and negative query-segment pairs. Experiments demonstrate that RAM outperforms existing baselines on multiple question answering and summarization benchmarks across two backbones, while delivering up to a 12x end-to-end speedup on long inputs (average length 16K; maximum length 32K). --- ### DOGMA: Weaving Structural Information into Data-centric Single-cell Transcriptomics Analysis Authors: Ru Zhang, Xunkai Li, Yaxin Deng, Sicheng Liu, Daohan Su, Qiangqiang Dai, Hongchao Qin, Rong-Hua Li, Guoren Wang, Jia Li Published: 2026-02-02 Categories: cs.LG, cs.AI, q-bio.GN arXiv: https://arxiv.org/abs/2602.01839v2 Recently, data-centric AI methodology has been a dominant paradigm in single-cell transcriptomics analysis, which treats data representation rather than model complexity as the fundamental bottleneck. In the review of current studies, earlier sequence methods treat cells as independent entities and adapt prevalent ML models to analyze their directly inherited sequence data. Despite their simplicity and intuition, these methods overlook the latent intercellular relationships driven by the functional mechanisms of biological systems and the inherent quality issues of the raw sequencing data. Therefore, a series of structured methods has emerged. Although they employ various heuristic rules to capture intricate intercellular relationships and enhance the raw sequencing data, these methods often neglect biological prior knowledge. This omission incurs substantial overhead and yields suboptimal graph representations, hindering the utility of ML models. To address these issues, we propose DOGMA, a data-centric framework designed for the structural reshaping and semantic enhancement of raw data through multi-level biological prior knowledge. Transcending reliance on purely data-driven heuristics, DOGMA provides a prior-guided graph construction pipeline that integrates statistical alignment with Cell Ontology and phylogenetic structure for biologically grounded cell-graph construction and robust cross-species alignment. Furthermore, Gene Ontology is utilized to bridge the feature-level semantic gap by incorporating functional priors. In complex multi-species and multi-organ benchmarks, DOGMA exhibits strong robustness in strict zero-shot cell-type evaluation and sample efficiency while using substantially lower GPU memory and inference time in downstream evaluation. --- ### AXE: Low-Cost Cross-Domain Web Structured Information Extraction Authors: Abdelrahman Mansour, Khaled W. Alshaer, Moataz Elsaban Published: 2026-02-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.01838v2 Extracting structured data from the web is often a trade-off between the brittle nature of manual heuristics and the prohibitive cost of Large Language Models. We introduce AXE (Adaptive X-Path Extractor), a pipeline that rethinks this process by treating the HTML DOM as a tree that needs pruning rather than just a wall of text to be read. AXE uses a specialized "pruning" mechanism to strip away boilerplate and irrelevant nodes, leaving behind a distilled, high-density context that allows a tiny 0.6B LLM to generate precise, structured outputs. To keep the model honest, we implement Grounded XPath Resolution (GXR), ensuring every extraction is physically traceable to a source node. Despite its low footprint, AXE achieves state-of-the-art zero-shot performance, outperforming several much larger, fully-trained alternatives with an F1 score of 88.1% on the SWDE dataset. By releasing our specialized adaptors, we aim to provide a practical, cost-effective path for large-scale web information extraction. Our code and adaptors are publicly available at https://github.com/abdo-Mansour/axetract. --- ### Efficient Cross-Country Data Acquisition Strategy for ADAS via Street-View Imagery Authors: Yin Wu, Daniel Slieter, Carl Esselborn, Ahmed Abouelazm, Tsung Yuan Tseng, J. Marius Zöllner Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01836v1 Deploying ADAS and ADS across countries remains challenging due to differences in legislation, traffic infrastructure, and visual conventions, which introduce domain shifts that degrade perception performance. Traditional cross-country data collection relies on extensive on-road driving, making it costly and inefficient to identify representative locations. To address this, we propose a street-view-guided data acquisition strategy that leverages publicly available imagery to identify places of interest (POI). Two POI scoring methods are introduced: a KNN-based feature distance approach using a vision foundation model, and a visual-attribution approach using a vision-language model. To enable repeatable evaluation, we adopt a collect-detect protocol and construct a co-located dataset by pairing the Zenseact Open Dataset with Mapillary street-view images. Experiments on traffic sign detection, a task particularly sensitive to cross-country variations in sign appearance, show that our approach achieves performance comparable to random sampling while using only half of the target-domain data. We further provide cost estimations for full-country analysis, demonstrating that large-scale street-view processing remains economically feasible. These results highlight the potential of street-view-guided data acquisition for efficient and cost-effective cross-country model adaptation. --- ### Synesthesia of Vehicles: Tactile Data Synthesis from Visual Inputs Authors: Rui Wang, Yaoguang Cao, Yuyi Chen, Jianyi Xu, Zhuoyang Li, Jiachen Shang, Shichun Yang Published: 2026-02-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01832v1 Autonomous vehicles (AVs) rely on multi-modal fusion for safety, but current visual and optical sensors fail to detect road-induced excitations which are critical for vehicles' dynamic control. Inspired by human synesthesia, we propose the Synesthesia of Vehicles (SoV), a novel framework to predict tactile excitations from visual inputs for autonomous vehicles. We develop a cross-modal spatiotemporal alignment method to address temporal and spatial disparities. Furthermore, a visual-tactile synesthetic (VTSyn) generative model using latent diffusion is proposed for unsupervised high-quality tactile data synthesis. A real-vehicle perception system collected a multi-modal dataset across diverse road and lighting conditions. Extensive experiments show that VTSyn outperforms existing models in temporal, frequency, and classification performance, enhancing AV safety through proactive tactile perception. --- ### Hyperbolic Graph Neural Networks Under the Microscope: The Role of Geometry-Task Alignment Authors: Dionisia Naddeo, Jonas Linkerhägner, Nicola Toschi, Geri Skenderi, Veronica Lachi Published: 2026-02-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01828v2 Many complex networks exhibit hierarchical, tree-like structures, making hyperbolic space a natural candidate wherein to learn representations of them. Based on this observation, Hyperbolic Graph Neural Networks (HGNNs) have been widely adopted as a principled choice for representation learning on tree-like graphs. In this work, we question this paradigm by proposing the additional condition of geometry--task alignment, i.e., whether the metric structure of the target follows that of the input graph. We theoretically and empirically demonstrate the capability of HGNNs to recover low-distortion representations on regression problems, and show that their geometric inductive bias becomes helpful when the problem requires preserving metric structure. By jointly analyzing predictive performance and embedding distortion, we further show that HGNNs gain an advantage on link prediction, a naturally geometry-aligned task, whereas this advantage largely disappears on standard node classification benchmarks, which are typically not geometry--aligned. Overall, our findings shift the focus from only asking "Is the graph hyperbolic?" to also questioning "Is the task aligned with hyperbolic geometry?", showing that HGNNs consistently outperform Euclidean models under such alignment, while their advantage vanishes otherwise. --- ### Beyond Precision: Training-Inference Mismatch is an Optimization Problem and Simple LR Scheduling Fixes It Authors: Yaxiang Zhang, Yingru Li, Jiacai Liu, Jiawei Xu, Ziniu Li, Qian Liu, Haoyuan Li Published: 2026-02-02 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.01826v1 Reinforcement Learning (RL) for training Large Language Models is notoriously unstable. While recent studies attribute this to "training inference mismatch stemming" from inconsistent hybrid engines, standard remedies, such as Importance Sampling, might fail during extended training runs. In this work, we analyze this instability through the lens of optimization, demonstrating that gradient noise and training-inference mismatch escalate in tandem as training progresses. Meanwhile, we find that the mismatch can be effectively suppressed by shrinking the update size. Taken together, we deduce that the mismatch is not merely a static numerical discrepancy, but a dynamic failure coupled with the model's optimization. Based on this insight, we propose a simple yet effective solution: a specialized Learning Rate (LR) scheduler. Instead of pre-defined decay schedule in traditional LR scheduler, our method dynamically triggers LR decay based on response length, which we identify as a reliable early-warning signal for impending instability. Empirical evidence suggests that by reducing the learning rate as gradient noise rises, we can consistently stabilize RL training and keep the training-inference mismatch at a safe level. --- ### Learning Sequential Decisions from Multiple Sources via Group-Robust Markov Decision Processes Authors: Mingyuan Xu, Zongqi Xia, Tianxi Cai, Doudou Zhou, Nian Si Published: 2026-02-02 Categories: stat.ME, cs.LG, math.OC, stat.ML arXiv: https://arxiv.org/abs/2602.01825v1 We often collect data from multiple sites (e.g., hospitals) that share common structure but also exhibit heterogeneity. This paper aims to learn robust sequential decision-making policies from such offline, multi-site datasets. To model cross-site uncertainty, we study distributionally robust MDPs with a group-linear structure: all sites share a common feature map, and both the transition kernels and expected reward functions are linear in these shared features. We introduce feature-wise (d-rectangular) uncertainty sets, which preserve tractable robust Bellman recursions while maintaining key cross-site structure. Building on this, we then develop an offline algorithm based on pessimistic value iteration that includes: (i) per-site ridge regression for Bellman targets, (ii) feature-wise worst-case (row-wise minimization) aggregation, and (iii) a data-dependent pessimism penalty computed from the diagonals of the inverse design matrices. We further propose a cluster-level extension that pools similar sites to improve sample efficiency, guided by prior knowledge of site similarity. Under a robust partial coverage assumption, we prove a suboptimality bound for the resulting policy. Overall, our framework addresses multi-site learning with heterogeneous data sources and provides a principled approach to robust planning without relying on strong state-action rectangularity assumptions. --- ### Seeing Is Believing? A Benchmark for Multimodal Large Language Models on Visual Illusions and Anomalies Authors: Wenjin Hou, Wei Liu, Han Hu, Xiaoxiao Sun, Serena Yeung-Levy, Hehe Fan Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01816v1 Multimodal Large Language Models (MLLMs) have shown remarkable proficiency on general-purpose vision-language benchmarks, reaching or even exceeding human-level performance. However, these evaluations typically rely on standard in-distribution data, leaving the robustness of MLLMs largely unexamined when faced with scenarios that defy common-sense priors. To address this gap, we introduce VIA-Bench, a challenging benchmark designed to probe model performance on visual illusions and anomalies. It includes six core categories: color illusions, motion illusions, gestalt illusions, geometric and spatial illusions, general visual illusions, and visual anomalies. Through careful human-in-the-loop review, we construct over 1K high-quality question-answer pairs that require nuanced visual reasoning. Extensive evaluation of over 20 state-of-the-art MLLMs, including proprietary, open-source, and reasoning-enhanced models, uncovers significant vulnerabilities. Notably, we find that Chain-of-Thought (CoT) reasoning offers negligible robustness, often yielding ``brittle mirages'' where the model's logic collapses under illusory stimuli. Our findings reveal a fundamental divergence between machine and human perception, suggesting that resolving such perceptual bottlenecks is critical for the advancement of artificial general intelligence. The benchmark data and code will be released. --- ### INDIBATOR: Diverse and Fact-Grounded Individuality for Multi-Agent Debate in Molecular Discovery Authors: Yunhui Jang, Seonghyun Park, Jaehyung Kim, Sungsoo Ahn Published: 2026-02-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01815v1 Multi-agent systems have emerged as a powerful paradigm for automating scientific discovery. To differentiate agent behavior in the multi-agent system, current frameworks typically assign generic role-based personas such as ''reviewer'' or ''writer'' or rely on coarse grained keyword-based personas. While functional, this approach oversimplifies how human scientists operate, whose contributions are shaped by their unique research trajectories. In response, we propose INDIBATOR, a framework for molecular discovery that grounds agents in individualized scientist profiles constructed from two modalities: publication history for literature-derived knowledge and molecular history for structural priors. These agents engage in multi-turn debate through proposal, critique, and voting phases. Our evaluation demonstrates that these fine-grained individuality-grounded agents consistently outperform systems relying on coarse-grained personas, achieving competitive or state-of-the-art performance. These results validate that capturing the ``scientific DNA'' of individual agents is essential for high-quality discovery. --- ### GPD: Guided Progressive Distillation for Fast and High-Quality Video Generation Authors: Xiao Liang, Yunzhu Zhang, Linchao Zhu Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01814v1 Diffusion models have achieved remarkable success in video generation; however, the high computational cost of the denoising process remains a major bottleneck. Existing approaches have shown promise in reducing the number of diffusion steps, but they often suffer from significant quality degradation when applied to video generation. We propose Guided Progressive Distillation (GPD), a framework that accelerates the diffusion process for fast and high-quality video generation. GPD introduces a novel training strategy in which a teacher model progressively guides a student model to operate with larger step sizes. The framework consists of two key components: (1) an online-generated training target that reduces optimization difficulty while improving computational efficiency, and (2) frequency-domain constraints in the latent space that promote the preservation of fine-grained details and temporal dynamics. Applied to the Wan2.1 model, GPD reduces the number of sampling steps from 48 to 6 while maintaining competitive visual quality on VBench. Compared with existing distillation methods, GPD demonstrates clear advantages in both pipeline simplicity and quality preservation. --- ### LDRNet: Large Deformation Registration Model for Chest CT Registration Authors: Cheng Wang, Qiyu Gao, Fandong Zhang, Shu Zhang, Yizhou Yu Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01812v1 Most of the deep learning based medical image registration algorithms focus on brain image registration tasks.Compared with brain registration, the chest CT registration has larger deformation, more complex background and region over-lap. In this paper, we propose a fast unsupervised deep learning method, LDRNet, for large deformation image registration of chest CT images. We first predict a coarse resolution registration field, then refine it from coarse to fine. We propose two innovative technical components: 1) a refine block that is used to refine the registration field in different resolutions, 2) a rigid block that is used to learn transformation matrix from high-level features. We train and evaluate our model on the private dataset and public dataset SegTHOR. We compare our performance with state-of-the-art traditional registration methods as well as deep learning registration models VoxelMorph, RCN, and LapIRN. The results demonstrate that our model achieves state-of-the-art performance for large deformation images registration and is much faster. --- ### Sentence Curve Language Models Authors: DongNyeong Heo, Taehwan Kim, Heeyoul Choi Published: 2026-02-02 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.01807v3 Language models (LMs) are a central component of modern AI systems, and diffusion language models (DLMs) have recently emerged as a competitive alternative. Both paradigms rely on word embeddings not only to represent the input sentence, but also to represent the target sentence that backbone models are trained to predict. We argue that such static embedding of the target word is insensitive to neighboring words, encouraging locally accurate word prediction while global sentence structure is less emphasized. To address this, we propose a continuous sentence representation, termed sentence curve, defined as a spline curve whose control points affect multiple words in the sentence. Based on this representation, we introduce sentence curve language model (SCLM), which extends DLMs to predict sentence curves instead of the static word embeddings. We theoretically show that sentence curve prediction induces a regularization effect that promotes global structure modeling, and characterize how different sentence curve types affect this behavior. Empirically, SCLM achieves state-of-the-art performance among DLMs on IWSLT14 and WMT14, shows stable training without burdensome knowledge distillation, and demonstrates promising potential compared to discrete DLMs on LM1B. --- ### FlowBypass: Rectified Flow Trajectory Bypass for Training-Free Image Editing Authors: Menglin Han, Zhangkai Ni Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01805v1 Training-free image editing has attracted increasing attention for its efficiency and independence from training data. However, existing approaches predominantly rely on inversion-reconstruction trajectories, which impose an inherent trade-off: longer trajectories accumulate errors and compromise fidelity, while shorter ones fail to ensure sufficient alignment with the edit prompt. Previous attempts to address this issue typically employ backbone-specific feature manipulations, limiting general applicability. To address these challenges, we propose FlowBypass, a novel and analytical framework grounded in Rectified Flow that constructs a bypass directly connecting inversion and reconstruction trajectories, thereby mitigating error accumulation without relying on feature manipulations. We provide a formal derivation of two trajectories, from which we obtain an approximate bypass formulation and its numerical solution, enabling seamless trajectory transitions. Extensive experiments demonstrate that FlowBypass consistently outperforms state-of-the-art image editing methods, achieving stronger prompt alignment while preserving high-fidelity details in irrelevant regions. --- ### Fast Autoregressive Video Diffusion and World Models with Temporal Cache Compression and Sparse Attention Authors: Dvir Samuel, Issar Tzachor, Matan Levy, Michael Green, Gal Chechik, Rami Ben-Ari Published: 2026-02-02 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.01801v2 Autoregressive video diffusion models enable streaming generation, opening the door to long-form synthesis, video world models, and interactive neural game engines. However, their core attention layers become a major bottleneck at inference time: as generation progresses, the KV cache grows, causing both increasing latency and escalating GPU memory, which in turn restricts usable temporal context and harms long-range consistency. In this work, we study redundancy in autoregressive video diffusion and identify three persistent sources: near-duplicate cached keys across frames, slowly evolving (largely semantic) queries/keys that make many attention computations redundant, and cross-attention over long prompts where only a small subset of tokens matters per frame. Building on these observations, we propose a unified, training-free attention framework (FAST-AR) for FAST-AutoRegressive diffusion, consisting of three components: TempCache compresses the KV cache via temporal correspondence to bound cache growth; AnnCA accelerates cross-attention by selecting frame-relevant prompt tokens using fast approximate nearest neighbor (ANN) matching; and AnnSA sparsifies self-attention by restricting each query to semantically matched keys, also using a lightweight ANN. Together, these modules reduce attention, compute, and memory and are compatible with existing autoregressive diffusion backbones and world models. Experiments demonstrate up to x5 - x10 end-to-end speedups while preserving near-identical visual quality and, crucially, maintaining stable throughput and nearly constant peak GPU memory usage over long rollouts, where prior methods progressively slow down and suffer from increasing memory usage. --- ### Spatio-Temporal Transformers for Long-Term NDVI Forecasting Authors: Ido Faran, Nathan S. Netanyahu, Maxim Shoshany Published: 2026-02-02 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2602.01799v1 Long-term satellite image time series (SITS) analysis in heterogeneous landscapes faces significant challenges, particularly in Mediterranean regions where complex spatial patterns, seasonal variations, and multi-decade environmental changes interact across different scales. This paper presents the Spatio-Temporal Transformer for Long Term Forecasting (STT-LTF ), an extended framework that advances beyond purely temporal analysis to integrate spatial context modeling with temporal sequence prediction. STT-LTF processes multi-scale spatial patches alongside temporal sequences (up to 20 years) through a unified transformer architecture, capturing both local neighborhood relationships and regional climate influences. The framework employs comprehensive self-supervised learning with spatial masking, temporal masking, and horizon sampling strategies, enabling robust model training from 40 years of unlabeled Landsat imagery. Unlike autoregressive approaches, STT-LTF directly predicts arbitrary future time points without error accumulation, incorporating spatial patch embeddings, cyclical temporal encoding, and geographic coordinates to learn complex dependencies across heterogeneous Mediterranean ecosystems. Experimental evaluation on Landsat data (1984-2024) demonstrates that STT-LTF achieves a Mean Absolute Error (MAE) of 0.0328 and R^2 of 0.8412 for next-year predictions, outperforming traditional statistical methods, CNN-based approaches, LSTM networks, and standard transformers. The framework's ability to handle irregular temporal sampling and variable prediction horizons makes it particularly suitable for analysis of heterogeneous landscapes experiencing rapid ecological transitions. --- ### ORCH: many analyses, one merge-a deterministic multi-agent orchestrator for discrete-choice reasoning with EMA-guided routing Authors: Hanlin Zhou, Huah Yong Chan Published: 2026-02-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01797v1 Recent advances in large-scale language models (LLMs) have made multi-agent architectures attractive for challenging reasoning tasks. However, many existing systems rely on stochastic routing or ad-hoc heuristics, making their behavior difficult to reproduce and their decision process hard to interpret. We propose ORCH, a deterministic coordination framework for discrete-choice reasoning that orchestrates heterogeneous LLMs. ORCH follows a ``many analyses, one decision'' paradigm: multiple base models independently produce structured analyses, and a dedicated merge agent outputs the final choice. The framework uses fixed rules for task decomposition and answer aggregation, keeping the pipeline predictable, reproducible, and training-free. Determinism here refers to fixed routing and aggregation rules under a fixed evaluation protocol, rather than strict bit-level reproducibility across deployments. To exploit model complementarity, we optionally introduce an EMA-guided router that updates agent selection using historical accuracy, latency, or cost; since it relies on answer-based feedback, it is mainly intended for benchmarking, controlled evaluation, or delayed-feedback settings. Experiments on MMLU, MMLU-Pro, and GSM8K show that ORCH consistently outperforms single-model baselines and a majority-vote ensemble. On MMLU-Pro, ORCH improves accuracy by over 10 points compared to the strongest baseline, and on GSM8K it yields gains exceeding 50 points; McNemar tests confirm statistical significance. The EMA router provides an additional 0.7--2.0 point accuracy boost, and ablations show that both multi-agent collaboration and routing contribute substantially. Overall, ORCH offers a practical path toward controllable, interpretable, and deployment-ready LLM-based agent systems for discrete-choice reasoning. --- ### RedVisor: Reasoning-Aware Prompt Injection Defense via Zero-Copy KV Cache Reuse Authors: Mingrui Liu, Sixiao Zhang, Cheng Long, Kwok-Yan Lam Published: 2026-02-02 Categories: cs.CR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.01795v1 Large Language Models (LLMs) are increasingly vulnerable to Prompt Injection (PI) attacks, where adversarial instructions hidden within retrieved contexts hijack the model's execution flow. Current defenses typically face a critical trade-off: prevention-based fine-tuning often degrades general utility via the "alignment tax", while detection-based filtering incurs prohibitive latency and memory costs. To bridge this gap, we propose RedVisor, a unified framework that synthesizes the explainability of detection systems with the seamless integration of prevention strategies. To the best of our knowledge, RedVisor is the first approach to leverage fine-grained reasoning paths to simultaneously detect attacks and guide the model's safe response. We implement this via a lightweight, removable adapter positioned atop the frozen backbone. This adapter serves a dual function: it first generates an explainable analysis that precisely localizes the injection and articulates the threat, which then explicitly conditions the model to reject the malicious command. Uniquely, the adapter is active only during this reasoning phase and is effectively muted during the subsequent response generation. This architecture yields two distinct advantages: (1) it mathematically preserves the backbone's original utility on benign inputs; and (2) it enables a novel KV Cache Reuse strategy, eliminating the redundant prefill computation inherent to decoupled pipelines. We further pioneer the integration of this defense into the vLLM serving engine with custom kernels. Experiments demonstrate that RedVisor outperforms state-of-the-art defenses in detection accuracy and throughput while incurring negligible utility loss. --- ### Grad2Reward: From Sparse Judgment to Dense Rewards for Improving Open-Ended LLM Reasoning Authors: Zheng Zhang, Ao Lu, Yuanhao Zeng, Ziwei Shan, Jinjin Guo, Lufei Li, Yexin Li, Kan Ren Published: 2026-02-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01791v1 Reinforcement Learning with Verifiable Rewards (RLVR) has catalyzed significant breakthroughs in complex LLM reasoning within verifiable domains, such as mathematics and programming. Recent efforts have sought to extend this paradigm to open-ended tasks by employing LLMs-as-a-Judge to provide sequence-level rewards for policy optimization. However, these rewards are inherently sparse, failing to provide the fine-grained supervision necessary for generating complex, long-form trajectories. Furthermore, current work treats the Judge as a black-box oracle, discarding the rich intermediate feedback signals encoded in it. To address these limitations, we introduce Grad2Reward, a novel framework that extracts dense process rewards directly from the Judge's model inference process via a single backward pass. By leveraging gradient-based attribution, Grad2Reward enables precise token-level credit assignment, substantially enhancing training efficiency and reasoning quality. Additionally, Grad2Reward introduces a self-judging mechanism, allowing the policy to improve through its own evaluative signals without training specialized reward models or reliance on superior external Judges. The experiments demonstrate that policies optimized with Grad2Reward achieve outstanding performance across diverse open-ended tasks, affirming its effectiveness and broad generalizability. --- ### Seeing is Coding: On the Effectiveness of Vision Language Models in Code Understanding Authors: Yuling Shi, Chaoxiang Xie, Zhensu Sun, Yeheng Chen, Chenxu Zhang, Longfei Yun, Chengcheng Wan, Hongyu Zhang, David Lo, Xiaodong Gu Published: 2026-02-02 Categories: cs.CL, cs.SE arXiv: https://arxiv.org/abs/2602.01785v3 Large Language Models (LLMs) have achieved remarkable success in source code understanding, yet as software systems grow in scale, computational efficiency has become a critical bottleneck. Currently, these models rely on a text-based paradigm that treats source code as a linear sequence of tokens, which leads to a linear increase in context length and associated computational costs. The rapid advancement of Multimodal LLMs (MLLMs) introduces an opportunity to optimize efficiency by representing source code as rendered images. Unlike text, which is difficult to compress without losing semantic meaning, the image modality is inherently suitable for compression. By adjusting resolution, images can be scaled to a fraction of their original token cost while remaining recognizable to vision-capable models. To explore the feasibility of this approach, we conduct the first systematic study on the effectiveness of MLLMs for code understanding. Our experiments reveal that: (1) MLLMs can effectively understand code with substantial token reduction, achieving up to 8x compression; (2) MLLMs can effectively leverage visual cues such as syntax highlighting, improving code completion performance under 4x compression; and (3) Code-understanding tasks like clone detection exhibit exceptional resilience to visual compression, with some compression ratios even slightly outperforming raw text inputs. Our findings highlight both the potential and current limitations of MLLMs in code understanding, which points out a shift toward image-modality code representation as a pathway to more efficient inference. --- ### CodeOCR: On the Effectiveness of Vision Language Models in Code Understanding Authors: Yuling Shi, Chaoxiang Xie, Zhensu Sun, Yeheng Chen, Chenxu Zhang, Longfei Yun, Chengcheng Wan, Hongyu Zhang, David Lo, Xiaodong Gu Published: 2026-02-02 Categories: cs.CL, cs.SE arXiv: https://arxiv.org/abs/2602.01785v2 Large Language Models (LLMs) have achieved remarkable success in source code understanding, yet as software systems grow in scale, computational efficiency has become a critical bottleneck. Currently, these models rely on a text-based paradigm that treats source code as a linear sequence of tokens, which leads to a linear increase in context length and associated computational costs. The rapid advancement of Multimodal LLMs (MLLMs) introduces an opportunity to optimize efficiency by representing source code as rendered images. Unlike text, which is difficult to compress without losing semantic meaning, the image modality is inherently suitable for compression. By adjusting resolution, images can be scaled to a fraction of their original token cost while remaining recognizable to vision-capable models. To explore the feasibility of this approach, we conduct the first systematic study on the effectiveness of MLLMs for code understanding. Our experiments reveal that: (1) MLLMs can effectively understand code with substantial token reduction, achieving up to 8x compression; (2) MLLMs can effectively leverage visual cues such as syntax highlighting, improving code completion performance under 4x compression; and (3) Code-understanding tasks like clone detection exhibit exceptional resilience to visual compression, with some compression ratios even slightly outperforming raw text inputs. Our findings highlight both the potential and current limitations of MLLMs in code understanding, which points out a shift toward image-modality code representation as a pathway to more efficient inference. --- ### Automated Discontinuity Set Characterisation in Enclosed Rock Face Point Clouds Using Single-Shot Filtering and Cyclic Orientation Transformation Authors: Dibyayan Patra, Pasindu Ranasinghe, Bikram Banerjee, Simit Raval Published: 2026-02-02 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.01783v1 Characterisation of structural discontinuity sets in exposed rock faces of underground mine cavities is essential for assessing rock-mass stability, excavation safety, and operational efficiency. UAV and other mobile laser-scanning techniques provide efficient means of collecting point clouds from rock faces. However, the development of a robust and efficient approach for automatic characterisation of discontinuity sets in real-world scenarios, like fully enclosed rock faces in cavities, remains an open research problem. In this study, a new approach is proposed for automatic discontinuity set characterisation that uses a single-shot filtering strategy, an innovative cyclic orientation transformation scheme and a hierarchical clustering technique. The single-shot filtering step isolates planar regions while robustly suppressing noise and high-curvature artefacts in one pass using a signal-processing technique. To address the limitations of Cartesian clustering on polar orientation data, a cyclic orientation transformation scheme is developed, enabling accurate representation of dip angle and dip direction in Cartesian space. The transformed orientations are then characterised into sets using a hierarchical clustering technique, which handles varying density distributions and identifies clusters without requiring user-defined set numbers. The accuracy of the method is validated on real-world mine stope and against ground truth obtained using manually handpicked discontinuity planes identified with the Virtual Compass tool, as well as widely used automated structure mapping techniques. The proposed approach outperforms the other techniques by exhibiting the lowest mean absolute error in estimating discontinuity set orientations in real-world stope data with errors of 1.95° and 2.20° in nominal dip angle and dip direction, respectively, and dispersion errors lying below 3°. --- ### DDP-WM: Disentangled Dynamics Prediction for Efficient World Models Authors: Shicheng Yin, Kaixuan Yin, Weixing Chen, Yang Liu, Guanbin Li, Liang Lin Published: 2026-02-02 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2602.01780v4 World models are essential for autonomous robotic planning. However, the substantial computational overhead of existing dense Transformerbased models significantly hinders real-time deployment. To address this efficiency-performance bottleneck, we introduce DDP-WM, a novel world model centered on the principle of Disentangled Dynamics Prediction (DDP). We hypothesize that latent state evolution in observed scenes is heterogeneous and can be decomposed into sparse primary dynamics driven by physical interactions and secondary context-driven background updates. DDP-WM realizes this decomposition through an architecture that integrates efficient historical processing with dynamic localization to isolate primary dynamics. By employing a crossattention mechanism for background updates, the framework optimizes resource allocation and provides a smooth optimization landscape for planners. Extensive experiments demonstrate that DDP-WM achieves significant efficiency and performance across diverse tasks, including navigation, precise tabletop manipulation, and complex deformable or multi-body interactions. Specifically, on the challenging Push-T task, DDP-WM achieves an approximately 9 times inference speedup and improves the MPC success rate from 90% to98% compared to state-of-the-art dense models. The results establish a promising path for developing efficient, high-fidelity world models. Codes is available at https://hcplab-sysu.github.io/DDP-WM/. --- ### LingLanMiDian: Systematic Evaluation of LLMs on TCM Knowledge and Clinical Reasoning Authors: Rui Hua, Yu Wei, Zixin Shu, Kai Chang, Dengying Yan, Jianan Xia, Zeyu Liu, Hui Zhu, Shujie Song, Mingzhong Xiao, Xiaodong Li, Dongmei Jia et al. Published: 2026-02-02 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.01779v1 Large language models (LLMs) are advancing rapidly in medical NLP, yet Traditional Chinese Medicine (TCM) with its distinctive ontology, terminology, and reasoning patterns requires domain-faithful evaluation. Existing TCM benchmarks are fragmented in coverage and scale and rely on non-unified or generation-heavy scoring that hinders fair comparison. We present the LingLanMiDian (LingLan) benchmark, a large-scale, expert-curated, multi-task suite that unifies evaluation across knowledge recall, multi-hop reasoning, information extraction, and real-world clinical decision-making. LingLan introduces a consistent metric design, a synonym-tolerant protocol for clinical labels, a per-dataset 400-item Hard subset, and a reframing of diagnosis and treatment recommendation into single-choice decision recognition. We conduct comprehensive, zero-shot evaluations on 14 leading open-source and proprietary LLMs, providing a unified perspective on their strengths and limitations in TCM commonsense knowledge understanding, reasoning, and clinical decision support; critically, the evaluation on Hard subset reveals a substantial gap between current models and human experts in TCM-specialized reasoning. By bridging fundamental knowledge and applied reasoning through standardized evaluation, LingLan establishes a unified, quantitative, and extensible foundation for advancing TCM LLMs and domain-specific medical AI research. All evaluation data and code are available at https://github.com/TCMAI-BJTU/LingLan and http://tcmnlp.com. --- ### Data Distribution Matters: A Data-Centric Perspective on Context Compression for Large Language Model Authors: Kangtao Lv, Jiwei Tang, Langming Liu, Haibin Chen, Weidong Zhang, Shilei Liu, Yongwei Wang, Yujin Yuan, Wenbo Su, Bo Zheng Published: 2026-02-02 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.01778v1 The deployment of Large Language Models (LLMs) in long-context scenarios is hindered by computational inefficiency and significant information redundancy. Although recent advancements have widely adopted context compression to address these challenges, existing research only focus on model-side improvements, the impact of the data distribution itself on context compression remains largely unexplored. To bridge this gap, we are the first to adopt a data-centric perspective to systematically investigate how data distribution impacts compression quality, including two dimensions: input data and intrinsic data (i.e., the model's internal pretrained knowledge). We evaluate the semantic integrity of compressed representations using an autoencoder-based framework to systematically investigate it. Our experimental results reveal that: (1) encoder-measured input entropy negatively correlates with compression quality, while decoder-measured entropy shows no significant relationship under a frozen-decoder setting; and (2) the gap between intrinsic data of the encoder and decoder significantly diminishes compression gains, which is hard to mitigate. Based on these findings, we further present practical guidelines to optimize compression gains. --- ### Stein-Rule Shrinkage for Stochastic Gradient Estimation in High Dimensions Authors: M. Arashi, M. Amintoosi Published: 2026-02-02 Categories: cs.LG, cs.AI, math.ST, stat.ML arXiv: https://arxiv.org/abs/2602.01777v2 Stochastic gradient methods are central to large-scale learning, but they treat mini-batch gradients as unbiased estimators, which classical decision theory shows are inadmissible in high dimensions. We formulate gradient computation as a high-dimensional estimation problem and introduce a framework based on Stein-rule shrinkage. We construct a gradient estimator that adaptively contracts noisy mini-batch gradients toward a stable estimator derived from historical momentum. The shrinkage intensity is determined in a data-driven manner using an online estimate of gradient noise variance, leveraging statistics from adaptive optimizers. Under a Gaussian noise model, we show our estimator uniformly dominates the standard stochastic gradient under squared error loss and is minimax-optimal. We incorporate this into the Adam optimizer, yielding SR-Adam, a practical algorithm with negligible computational cost. Empirical evaluations on CIFAR10 and CIFAR100 across multiple levels of input noise show consistent improvements over Adam in the large-batch regime. Ablation studies indicate that gains arise primarily from selectively applying shrinkage to high-dimensional convolutional layers, while indiscriminate shrinkage across all parameters degrades performance. These results illustrate that classical shrinkage principles provide a principled approach to improving stochastic gradient estimation in deep learning. --- ### Position: Beyond Model-Centric Prediction -- Agentic Time Series Forecasting Authors: Mingyue Cheng, Xiaoyu Tao, Qi Liu, Ze Guo, Enhong Chen Published: 2026-02-02 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.01776v4 Time series forecasting has traditionally been formulated as a model-centric, static, and single-pass prediction problem that maps historical observations to future values. While this paradigm has driven substantial progress, it proves insufficient in adaptive and multi-turn settings where forecasting requires informative feature extraction, reasoning-driven inference, iterative refinement, and continual adaptation over time. In this paper, we argue for agentic time series forecasting (ATSF), which reframes forecasting as an agentic process composed of perception, planning, action, reflection, and memory. Rather than focusing solely on predictive models, ATSF emphasizes organizing forecasting as an agentic workflow that can interact with tools, incorporate feedback from outcomes, and evolve through experience accumulation. We outline three representative implementation paradigms -- workflow-based design, agentic reinforcement learning, and a hybrid agentic workflow paradigm -- and discuss the opportunities and challenges that arise when shifting from model-centric prediction to agentic forecasting. Together, this position aims to establish agentic forecasting as a foundation for future research at the intersection of time series forecasting. --- ### Efficient Cross-Architecture Knowledge Transfer for Large-Scale Online User Response Prediction Authors: Yucheng Wu, Yuekui Yang, Hongzheng Li, Anan Liu, Jian Xiao, Junjie Zhai, Huan Yu, Shaoping Ma, Leye Wang Published: 2026-02-02 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.01775v1 Deploying new architectures in large-scale user response prediction systems incurs high model switching costs due to expensive retraining on massive historical data and performance degradation under data retention constraints. Existing knowledge distillation methods struggle with architectural heterogeneity and the prohibitive cost of transferring large embedding tables. We propose CrossAdapt, a two-stage framework for efficient cross-architecture knowledge transfer. The offline stage enables rapid embedding transfer via dimension-adaptive projections without iterative training, combined with progressive network distillation and strategic sampling to reduce computational cost. The online stage introduces asymmetric co-distillation, where students update frequently while teachers update infrequently, together with a distribution-aware adaptation mechanism that dynamically balances historical knowledge preservation and fast adaptation to evolving data. Experiments on three public datasets show that CrossAdapt achieves 0.27-0.43% AUC improvements while reducing training time by 43-71%. Large-scale deployment on Tencent WeChat Channels (~10M daily samples) further demonstrates its effectiveness, significantly mitigating AUC degradation, LogLoss increase, and prediction bias compared to standard distillation baselines. --- ### Cost-Aware Bayesian Optimization for Prototyping Interactive Devices Authors: Thomas Langerak, Renate Zhang, Ziyuan Wang, Per Ola Kristensson, Antti Oulasvirta Published: 2026-02-02 Categories: cs.HC, cs.LG arXiv: https://arxiv.org/abs/2602.01774v1 Deciding which idea is worth prototyping is a central concern in iterative design. A prototype should be produced when the expected improvement is high and the cost is low. However, this is hard to decide, because costs can vary drastically: a simple parameter tweak may take seconds, while fabricating hardware consumes material and energy. Such asymmetries, can discourage a designer from exploring the design space. In this paper, we present an extension of cost-aware Bayesian optimization to account for diverse prototyping costs. The method builds on the power of Bayesian optimization and requires only a minimal modification to the acquisition function. The key idea is to use designer-estimated costs to guide sampling toward more cost-effective prototypes. In technical evaluations, the method achieved comparable utility to a cost-agnostic baseline while requiring only ${\approx}70\%$ of the cost; under strict budgets, it outperformed the baseline threefold. A within-subjects study with 12 participants in a realistic joystick design task demonstrated similar benefits. These results show that accounting for prototyping costs can make Bayesian optimization more compatible with real-world design projects. ---