# Research Papers — 2026-02 · bundle page 88 of 210 > Full-text markdown of papers #4351–4400 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/88 [← Previous bundle](https://aipod.dev/papers/archive/2026-02/page/87.md) [Next bundle →](https://aipod.dev/papers/archive/2026-02/page/89.md) ### Character-aware Transformers Learn an Irregular Morphological Pattern Yet None Generalize Like Humans Authors: Akhilesh Kakolu Ramarao, Kevin Tang, Dinah Baer-Henney Published: 2026-02-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.14100v1 Whether neural networks can serve as cognitive models of morphological learning remains an open question. Recent work has shown that encoder-decoder models can acquire irregular patterns, but evidence that they generalize these patterns like humans is mixed. We investigate this using the Spanish \emph{L-shaped morphome}, where only the first-person singular indicative (e.g., \textit{pongo} `I put') shares its stem with all subjunctive forms (e.g., \textit{ponga, pongas}) despite lacking apparent phonological, semantic, or syntactic motivation. We compare five encoder-decoder transformers varying along two dimensions: sequential vs. position-invariant positional encoding, and atomic vs. decomposed tag representations. Positional encoding proves decisive: position-invariant models recover the correct L-shaped paradigm clustering even when L-shaped verbs are scarce in training, whereas sequential positional encoding models only partially capture the pattern. Yet none of the models productively generalize this pattern to novel forms. Position-invariant models generalize the L-shaped stem across subjunctive cells but fail to extend it to the first-person singular indicative, producing a mood-based generalization rather than the L-shaped morphomic pattern. Humans do the opposite, generalizing preferentially to the first-person singular indicative over subjunctive forms. None of the models reproduce the human pattern, highlighting the gap between statistical pattern reproduction and morphological abstraction. --- ### SemanticFeels: Semantic Labeling during In-Hand Manipulation Authors: Anas Al Shikh Khalil, Haozhi Qi, Roberto Calandra Published: 2026-02-15 Categories: cs.RO, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2602.14099v1 As robots become increasingly integrated into everyday tasks, their ability to perceive both the shape and properties of objects during in-hand manipulation becomes critical for adaptive and intelligent behavior. We present SemanticFeels, an extension of the NeuralFeels framework that integrates semantic labeling with neural implicit shape representation, from vision and touch. To illustrate its application, we focus on material classification: high-resolution Digit tactile readings are processed by a fine-tuned EfficientNet-B0 convolutional neural network (CNN) to generate local material predictions, which are then embedded into an augmented signed distance field (SDF) network that jointly predicts geometry and continuous material regions. Experimental results show that the system achieves a high correspondence between predicted and actual materials on both single- and multi-material objects, with an average matching accuracy of 79.87% across multiple manipulation trials on a multi-material object. --- ### ForgeryVCR: Visual-Centric Reasoning via Efficient Forensic Tools in MLLMs for Image Forgery Detection and Localization Authors: Youqi Wang, Shen Chen, Haowei Wang, Rongxuan Peng, Taiping Yao, Shunquan Tan, Changsheng Chen, Bin Li, Shouhong Ding Published: 2026-02-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.14098v2 Existing Multimodal Large Language Models (MLLMs) for image forgery detection and localization predominantly operate under a text-centric Chain-of-Thought (CoT) paradigm. However, forcing these models to textually characterize imperceptible low-level tampering traces inevitably leads to hallucinations, as linguistic modalities are insufficient to capture such fine-grained pixel-level inconsistencies. To overcome this, we propose ForgeryVCR, a framework that incorporates a forensic toolbox to materialize imperceptible traces into explicit visual intermediates via Visual-Centric Reasoning. To enable efficient tool utilization, we introduce a Strategic Tool Learning post-training paradigm, encompassing gain-driven trajectory construction for Supervised Fine-Tuning (SFT) and subsequent Reinforcement Learning (RL) optimization guided by a tool utility reward. This paradigm empowers the MLLM to act as a proactive decision-maker, learning to spontaneously invoke multi-view reasoning paths including local zoom-in for fine-grained inspection and the analysis of invisible inconsistencies in compression history, noise residuals, and frequency domains. Extensive experiments reveal that ForgeryVCR achieves state-of-the-art (SOTA) performance in both detection and localization tasks, demonstrating superior generalization and robustness with minimal tool redundancy. The code is available at https://github.com/youqiwong/ForgeryVCR. --- ### ForgeryVCR: Visual-Centric Reasoning via Efficient Forensic Tools in MLLMs for Image Forgery Detection and Localization Authors: Youqi Wang, Shen Chen, Haowei Wang, Rongxuan Peng, Taiping Yao, Shunquan Tan, Changsheng Chen, Bin Li, Shouhong Ding Published: 2026-02-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.14098v1 Existing Multimodal Large Language Models (MLLMs) for image forgery detection and localization predominantly operate under a text-centric Chain-of-Thought (CoT) paradigm. However, forcing these models to textually characterize imperceptible low-level tampering traces inevitably leads to hallucinations, as linguistic modalities are insufficient to capture such fine-grained pixel-level inconsistencies. To overcome this, we propose ForgeryVCR, a framework that incorporates a forensic toolbox to materialize imperceptible traces into explicit visual intermediates via Visual-Centric Reasoning. To enable efficient tool utilization, we introduce a Strategic Tool Learning post-training paradigm, encompassing gain-driven trajectory construction for Supervised Fine-Tuning (SFT) and subsequent Reinforcement Learning (RL) optimization guided by a tool utility reward. This paradigm empowers the MLLM to act as a proactive decision-maker, learning to spontaneously invoke multi-view reasoning paths including local zoom-in for fine-grained inspection and the analysis of invisible inconsistencies in compression history, noise residuals, and frequency domains. Extensive experiments reveal that ForgeryVCR achieves state-of-the-art (SOTA) performance in both detection and localization tasks, demonstrating superior generalization and robustness with minimal tool redundancy. The project page is available at https://youqiwong.github.io/projects/ForgeryVCR/. --- ### NEST: Nascent Encoded Steganographic Thoughts Authors: Artem Karpov Published: 2026-02-15 Categories: cs.AI, cs.CR arXiv: https://arxiv.org/abs/2602.14095v2 Monitoring chain-of-thought (CoT) reasoning is a foundational safety technique for large language model agents; however, this oversight is compromised if models learn to conceal their reasoning. We explore steganographic CoT--where models hide secret reasoning within innocuous text--to inform risk assessment and deployment policies. Steganographic reasoning requires two skills in a single forward pass: computing an intermediate result, and embedding it into a coherent cover that answers an unrelated question. Drawing on our taxonomy of steganographic and non-steganographic CoT types, we systematically evaluate the limits of prompt-elicited steganographic CoT capability across 34 models, ranging from past generations to the current frontier. We measure monitor evasion, refusal rates, encoding fidelity, and hidden task accuracy across five datasets, comparing against plain reasoning, direct answer, and filler-token baselines. The two experiments isolate the two sub-skills: a reasoning tasks sweep tests joint reason-and-embed, while a counting task hands the model a known numerical sequence and tests embedding alone--a necessary precondition for stego reasoning. Current frontier models cannot sustain joint reason-and-embed: a paired McNemar comparison shows the steganographic channel is dominated by an filler-token baseline on every (model, family) cell. The encoding-only floor, by contrast, is cleared--Claude Opus~4.5 reaches 92% per-number partial accuracy on 4-digit sequences and saturates at 100% exact-match on length-8 single-digit sequences--establishing that the binding constraint on stego CoT is the joint reasoning-plus-encoding load, not raw channel capacity. Our findings underscore the need for continuous evaluation of steganographic risk and provide a methodology to preemptively detect and evaluate hidden reasoning that might empower misaligned scheming and deceptive behavior. --- ### GUI-GENESIS: Automated Synthesis of Efficient Environments with Verifiable Rewards for GUI Agent Post-Training Authors: Yuan Cao, Dezhi Ran, Mengzhou Wu, Yuzhe Guo, Xin Chen, Ang Li, Gang Cao, Gong Zhi, Hao Yu, Linyi Li, Wei Yang, Tao Xie Published: 2026-02-15 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.14093v1 Post-training GUI agents in interactive environments is critical for developing generalization and long-horizon planning capabilities. However, training on real-world applications is hindered by high latency, poor reproducibility, and unverifiable rewards relying on noisy visual proxies. To address the limitations, we present GUI-GENESIS, the first framework to automatically synthesize efficient GUI training environments with verifiable rewards. GUI-GENESIS reconstructs real-world applications into lightweight web environments using multimodal code models and equips them with code-native rewards, executable assertions that provide deterministic reward signals and eliminate visual estimation noise. Extensive experiments show that GUI-GENESIS reduces environment latency by 10 times and costs by over $28,000 per epoch compared to training on real applications. Notably, agents trained with GUI-GENESIS outperform the base model by 14.54% and even real-world RL baselines by 3.27% on held-out real-world tasks. Finally, we observe that models can synthesize environments they cannot yet solve, highlighting a pathway for self-improving agents. --- ### TabTracer: Monte Carlo Tree Search for Complex Table Reasoning with Large Language Models Authors: Zhizhao Luo, Zhaojing Luo, Meihui Zhang, Rui Mao Published: 2026-02-15 Categories: cs.DB, cs.AI arXiv: https://arxiv.org/abs/2602.14089v1 Large language models (LLMs) have emerged as powerful tools for natural language table reasoning, where there are two main categories of methods. Prompt-based approaches rely on language-only inference or one-pass program generation without step-level verification. Agent-based approaches use tools in a closed loop, but verification is often local and backtracking is limited, allowing errors to propagate and increasing cost. Moreover, they rely on chain- or beam-style trajectories that are typically combinatorially redundant, leading to high token costs. In this paper, we propose TabTracer, an agentic framework that coordinates multi-step tool calls over intermediate table states, with explicit state tracking for verification and rollback. First, it enforces step-level verification with typed operations and lightweight numeric and format checks to provide reliable rewards and suppress hallucinations. Second, execution-feedback Monte Carlo Tree Search maintains a search tree of candidate table states and uses backpropagated reflection scores to guide UCB1 selection and rollback via versioned snapshots. Third, it reduces redundancy with budget-aware pruning, deduplication, and state hashing with a monotonicity gate to cut token cost. Comprehensive evaluation on TabFact, WikiTQ, and CRT datasets shows that TabTracer outperforms state-of-the-art baselines by up to 6.7% in accuracy while reducing token consumption by 59--84%. --- ### Neural Optimal Transport in Hilbert Spaces: Characterizing Spurious Solutions and Gaussian Smoothing Authors: Jae-Hwan Choi, Jiwoo Yoon, Dohyun Kwon, Jaewoong Choi Published: 2026-02-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.14086v1 We study Neural Optimal Transport in infinite-dimensional Hilbert spaces. In non-regular settings, Semi-dual Neural OT often generates spurious solutions that fail to accurately capture target distributions. We analytically characterize this spurious solution problem using the framework of regular measures, which generalize Lebesgue absolute continuity in finite dimensions. To resolve ill-posedness, we extend the semi-dual framework via a Gaussian smoothing strategy based on Brownian motion. Our primary theoretical contribution proves that under a regular source measure, the formulation is well-posed and recovers a unique Monge map. Furthermore, we establish a sharp characterization for the regularity of smoothed measures, proving that the success of smoothing depends strictly on the kernel of the covariance operator. Empirical results on synthetic functional data and time-series datasets demonstrate that our approach effectively suppresses spurious solutions and outperforms existing baselines. --- ### Plan-MCTS: Plan Exploration for Action Exploitation in Web Navigation Authors: Weiming Zhang, Jihong Wang, Jiamu Zhou, Qingyao Li, Xinbei Ma, Congmin Zheng, Xingyu Lou, Weiwen Liu, Zhuosheng Zhang, Jun Wang, Yong Yu, Weinan Zhang Published: 2026-02-15 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.14083v1 Large Language Models (LLMs) have empowered autonomous agents to handle complex web navigation tasks. While recent studies integrate tree search to enhance long-horizon reasoning, applying these algorithms in web navigation faces two critical challenges: sparse valid paths that lead to inefficient exploration, and a noisy context that dilutes accurate state perception. To address this, we introduce Plan-MCTS, a framework that reformulates web navigation by shifting exploration to a semantic Plan Space. By decoupling strategic planning from execution grounding, it transforms sparse action space into a Dense Plan Tree for efficient exploration, and distills noisy contexts into an Abstracted Semantic History for precise state awareness. To ensure efficiency and robustness, Plan-MCTS incorporates a Dual-Gating Reward to strictly validate both physical executability and strategic alignment and Structural Refinement for on-policy repair of failed subplans. Extensive experiments on WebArena demonstrate that Plan-MCTS achieves state-of-the-art performance, surpassing current approaches with higher task effectiveness and search efficiency. --- ### CCiV: A Benchmark for Structure, Rhythm and Quality in LLM-Generated Chinese \textit{Ci} Poetry Authors: Shangqing Zhao, Yupei Ren, Yuhao Zhou, Xiaopeng Bai, Man Lan Published: 2026-02-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.14081v1 The generation of classical Chinese \textit{Ci} poetry, a form demanding a sophisticated blend of structural rigidity, rhythmic harmony, and artistic quality, poses a significant challenge for large language models (LLMs). To systematically evaluate and advance this capability, we introduce \textbf{C}hinese \textbf{Ci}pai \textbf{V}ariants (\textbf{CCiV}), a benchmark designed to assess LLM-generated \textit{Ci} poetry across these three dimensions: structure, rhythm, and quality. Our evaluation of 17 LLMs on 30 \textit{Cipai} reveals two critical phenomena: models frequently generate valid but unexpected historical variants of a poetic form, and adherence to tonal patterns is substantially harder than structural rules. We further show that form-aware prompting can improve structural and tonal control for stronger models, while potentially degrading weaker ones. Finally, we observe weak and inconsistent alignment between formal correctness and literary quality in our sample. CCiV highlights the need for variant-aware evaluation and more holistic constrained creative generation methods. --- ### Empty Shelves or Lost Keys? Recall Is the Bottleneck for Parametric Factuality Authors: Nitay Calderon, Eyal Ben-David, Zorik Gekhman, Eran Ofek, Gal Yona Published: 2026-02-15 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.14080v2 Standard factuality evaluations of LLMs treat all errors alike, obscuring whether failures arise from missing knowledge (empty shelves) or from limited access to encoded facts (lost keys). We propose a behavioral framework that profiles factual knowledge at the level of facts rather than questions, characterizing each fact by whether it is encoded, and then by how accessible it is: cannot be recalled, can be directly recalled, or can only be recalled with inference-time computation (thinking). To support such profiling, we introduce WikiProfile, a new benchmark constructed via an automated pipeline with a prompted LLM grounded in web search. Across 4 million responses from 13 LLMs, we find that encoding is nearly saturated in frontier models on our benchmark, with GPT-5 and Gemini-3 encoding 95--98% of facts. However, recall remains a major bottleneck: many errors previously attributed to missing knowledge instead stem from failures to access it. These failures are systematic and disproportionately affect long-tail facts and reverse questions. Finally, we show that thinking improves recall and can recover a substantial fraction of failures, indicating that future gains may rely less on scaling and more on methods that improve how models utilize what they already encode. --- ### Policy Gradient with Adaptive Entropy Annealing for Continual Fine-Tuning Authors: Yaqian Zhang, Bernhard Pfahringer, Eibe Frank, Albert Bifet Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14078v1 Despite their success, large pretrained vision models remain vulnerable to catastrophic forgetting when adapted to new tasks in class-incremental settings. Parameter-efficient fine-tuning (PEFT) alleviates this by restricting trainable parameters, yet most approaches still rely on cross-entropy (CE) loss, a surrogate for the 0-1 loss, to learn from new data. We revisit this choice and revive the true objective (0-1 loss) through a reinforcement learning perspective. By formulating classification as a one-step Markov Decision Process, we derive an Expected Policy Gradient (EPG) method that directly minimizes misclassification error with a low-variance gradient estimation. Our analysis shows that CE can be interpreted as EPG with an additional sample-weighting mechanism: CE encourages exploration by emphasizing low-confidence samples, while EPG prioritizes high-confidence ones. Building on this insight, we propose adaptive entropy annealing (aEPG), a training strategy that transitions from exploratory (CE-like) to exploitative (EPG-like) learning. aEPG-based methods outperform CE-based methods across diverse benchmarks and with various PEFT modules. More broadly, we evaluate various entropy regularization methods and demonstrate that lower entropy of the output prediction distribution enhances adaptation in pretrained vision models. --- ### GTS: Inference-Time Scaling of Latent Reasoning with a Learnable Gaussian Thought Sampler Authors: Minghan Wang, Ye Bai, Thuy-Trang Vu, Ehsan Shareghi, Gholamreza Haffari Published: 2026-02-15 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.14077v2 Inference-time scaling (ITS) in latent reasoning models typically relies on heuristic perturbations, such as dropout or fixed Gaussian noise, to generate diverse candidate trajectories. However, we show that stronger perturbations do not necessarily yield better sampling quality: they often induce larger distribution shifts without producing more useful reasoning paths or better final decisions. A key limitation is that these perturbations inject stochasticity without defining an explicit conditional sampling distribution, making latent exploration difficult to control or optimize. To address this, we propose the Gaussian Thought Sampler (GTS), a lightweight module that reformulates latent exploration as sampling from a learned conditional distribution over continuous reasoning states. GTS predicts context-dependent perturbation distributions and is trained with GRPO-style policy optimization while keeping the backbone frozen, turning heuristic perturbation into an explicit probabilistic sampling policy. Experiments across multiple benchmarks and two latent reasoning architectures show that GTS yields more reliable inference-time scaling than heuristic baselines, suggesting that effective latent ITS requires better-controlled and optimizable sampling rather than simply amplifying stochasticity. --- ### Annotation-Efficient Vision-Language Model Adaptation to the Polish Language Using the LLaVA Framework Authors: Grzegorz Statkiewicz, Alicja Dobrzeniecka, Karolina Seweryn, Aleksandra Krasnodębska, Karolina Piosek, Katarzyna Bogusz, Sebastian Cygert, Wojciech Kusa Published: 2026-02-15 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2602.14073v2 Most vision-language models (VLMs) are trained on English-centric data, limiting their performance in other languages and cultural contexts. This restricts their usability for non-English-speaking users and hinders the development of multimodal systems that reflect diverse linguistic and cultural realities. In this work, we reproduce and adapt the LLaVA-Next methodology to create a set of Polish VLMs. We rely on a fully automated pipeline for translating and filtering existing multimodal datasets, and complement this with synthetic Polish data for OCR and culturally specific tasks. Despite relying almost entirely on automatic translation and minimal manual intervention to the training data, our approach yields strong results: we observe a +9.5% improvement over LLaVA-1.6-Vicuna-13B on a Polish-adapted MMBench, along with higher-quality captions in generative evaluations, as measured by human annotators in terms of linguistic correctness. These findings highlight that large-scale automated translation, combined with lightweight filtering, can effectively bootstrap high-quality multimodal models for low-resource languages. Some challenges remain, particularly in cultural coverage and evaluation. To facilitate further research, we make our models and evaluation dataset publicly available. --- ### Bidirectional Temporal Dynamics Modeling for EEG-based Driving Fatigue Recognition Authors: Yip Tin Po, Jianming Wang, Yutao Miao, Jiayan Zhang, Yunxu Zhao, Xiaomin Ouyang, Zhihong Li, Nevin L. Zhang Published: 2026-02-15 Categories: cs.OH, cs.CV arXiv: https://arxiv.org/abs/2602.14071v3 Driving fatigue is a major contributor to traffic accidents and poses a serious threat to road safety. Electroencephalography (EEG) provides a direct measurement of neural activity, yet EEG-based fatigue recognition is hindered by strong non-stationarity and asymmetric neural dynamics. To address these challenges, we propose DeltaGateNet, a novel framework that explicitly captures Bidirectional temporal dynamics for EEG-based driving fatigue recognition. Our key idea is to introduce a Bidirectional Delta module that decomposes first-order temporal differences into positive and negative components, enabling explicit modeling of asymmetric neural activation and suppression patterns. Furthermore, we design a Gated Temporal Convolution module to capture long-term temporal dependencies for each EEG channel using depthwise temporal convolutions and residual learning, preserving channel-wise specificity while enhancing temporal representation robustness. Extensive experiments conducted under both intra-subject and inter-subject evaluation settings on the public SEED-VIG and SADT driving fatigue datasets demonstrate that DeltaGateNet consistently outperforms existing methods. On SEED-VIG, DeltaGateNet achieves an intra-subject accuracy of 81.89% and an inter-subject accuracy of 55.55%. On the balanced SADT 2022 dataset, it attains intra-subject and inter-subject accuracies of 96.81% and 83.21%, respectively, while on the unbalanced SADT 2952 dataset, it achieves 96.84% intra-subject and 84.49% inter-subject accuracy. These results indicate that explicitly modeling Bidirectional temporal dynamics yields robust and generalizable performance under varying subject and class-distribution conditions. --- ### Bidirectional Temporal Dynamics Modeling for EEG-based Driving Fatigue Recognition Authors: Yip Tin Po, Jianming Wang, Yutao Miao, Jiayan Zhang, Yunxu Zhao, Xiaomin Ouyang, Zhihong Li, Nevin L. Zhang Published: 2026-02-15 Categories: cs.OH, cs.CV arXiv: https://arxiv.org/abs/2602.14071v2 Driving fatigue is a major contributor to traffic accidents and poses a serious threat to road safety. Electroencephalography (EEG) provides a direct measurement of neural activity, yet EEG-based fatigue recognition is hindered by strong non-stationarity and asymmetric neural dynamics. To address these challenges, we propose DeltaGateNet, a novel framework that explicitly captures Bidirectional temporal dynamics for EEG-based driving fatigue recognition. Our key idea is to introduce a Bidirectional Delta module that decomposes first-order temporal differences into positive and negative components, enabling explicit modeling of asymmetric neural activation and suppression patterns. Furthermore, we design a Gated Temporal Convolution module to capture long-term temporal dependencies for each EEG channel using depthwise temporal convolutions and residual learning, preserving channel-wise specificity while enhancing temporal representation robustness. Extensive experiments conducted under both intra-subject and inter-subject evaluation settings on the public SEED-VIG and SADT driving fatigue datasets demonstrate that DeltaGateNet consistently outperforms existing methods. On SEED-VIG, DeltaGateNet achieves an intra-subject accuracy of 81.89% and an inter-subject accuracy of 55.55%. On the balanced SADT 2022 dataset, it attains intra-subject and inter-subject accuracies of 96.81% and 83.21%, respectively, while on the unbalanced SADT 2952 dataset, it achieves 96.84% intra-subject and 84.49% inter-subject accuracy. These results indicate that explicitly modeling Bidirectional temporal dynamics yields robust and generalizable performance under varying subject and class-distribution conditions. --- ### Open Rubric System: Scaling Reinforcement Learning with Pairwise Adaptive Rubric Authors: Ruipeng Jia, Yunyi Yang, Wen Wang, Yuxin Wu, Yongbo Gai, Siyuan Tao, Mengyu Zhou, Jianhe Lin, Xiaoxi Jiang, Guanjun Jiang Published: 2026-02-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.14069v3 Scalar reward models compress multi-dimensional human preferences into a single opaque score, creating an information bottleneck that often leads to brittleness and reward hacking in open-ended alignment. We argue that robust alignment for non-verifiable tasks is fundamentally a principle generalization problem: reward should not be a learned function internalized into a judge, but an explicit reasoning process executed under inspectable principles. To operationalize this view, we present the Open Rubric System (OpenRS), a plug-and-play, rubrics-based LLM-as-a-Judge framework built around Pairwise Adaptive Meta-Rubrics (PAMR) and lightweight Pointwise Verifiable Rubrics (PVRs), which provide both hard-constraint guardrails and verifiable reward components when ground-truth or programmatic checks are available. OpenRS uses an explicit meta-rubric -- a constitution-like specification that governs how rubrics are instantiated, weighted, and enforced -- and instantiates adaptive rubrics on the fly by conditioning on the semantic differences between two candidate responses. It then performs criterion-wise pairwise comparisons and aggregates criterion-level preferences externally, avoiding pointwise weighted scalarization while improving discriminability in open-ended settings. To keep principles consistent yet editable across various domains, we introduce a two-level meta-rubric refinement pipeline (automated evolutionary refinement for general principles and a reproducible human-in-the-loop procedure for domain principles), complemented with pointwise verifiable rubrics that act as both guardrails against degenerate behaviors and a source of verifiable reward for objective sub-tasks. Finally, we instantiate OpenRS as reward supervision in pairwise RL training. --- ### Open Rubric System: Scaling Reinforcement Learning with Pairwise Adaptive Rubric Authors: Ruipeng Jia, Yunyi Yang, Yuxin Wu, Yongbo Gai, Siyuan Tao, Mengyu Zhou, Jianhe Lin, Xiaoxi Jiang, Guanjun Jiang Published: 2026-02-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.14069v2 Scalar reward models compress multi-dimensional human preferences into a single opaque score, creating an information bottleneck that often leads to brittleness and reward hacking in open-ended alignment. We argue that robust alignment for non-verifiable tasks is fundamentally a principle generalization problem: reward should not be a learned function internalized into a judge, but an explicit reasoning process executed under inspectable principles. To operationalize this view, we present the Open Rubric System (OpenRS), a plug-and-play, rubrics-based LLM-as-a-Judge framework built around Pairwise Adaptive Meta-Rubrics (PAMR) and lightweight Pointwise Verifiable Rubrics (PVRs), which provide both hard-constraint guardrails and verifiable reward components when ground-truth or programmatic checks are available. OpenRS uses an explicit meta-rubric -- a constitution-like specification that governs how rubrics are instantiated, weighted, and enforced -- and instantiates adaptive rubrics on the fly by conditioning on the semantic differences between two candidate responses. It then performs criterion-wise pairwise comparisons and aggregates criterion-level preferences externally, avoiding pointwise weighted scalarization while improving discriminability in open-ended settings. To keep principles consistent yet editable across various domains, we introduce a two-level meta-rubric refinement pipeline (automated evolutionary refinement for general principles and a reproducible human-in-the-loop procedure for domain principles), complemented with pointwise verifiable rubrics that act as both guardrails against degenerate behaviors and a source of verifiable reward for objective sub-tasks. Finally, we instantiate OpenRS as reward supervision in pairwise RL training. --- ### CoCoEdit: Content-Consistent Image Editing via Region Regularized Reinforcement Learning Authors: Yuhui Wu, Chenxi Xie, Ruibin Li, Liyi Chen, Qiaosi Yi, Lei Zhang Published: 2026-02-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.14068v2 Image editing has achieved impressive results with the development of large-scale generative models. However, existing models mainly focus on the editing effects of intended objects and regions, often leading to unwanted changes in unintended regions. We present a post-training framework for Content-Consistent Editing (CoCoEdit) via region regularized reinforcement learning. We first augment existing editing datasets with refined instructions and masks, from which 40K diverse and high quality samples are curated as training set. We then introduce a pixel-level similarity reward to complement MLLM-based rewards, enabling models to ensure both editing quality and content consistency during the editing process. To overcome the spatial-agnostic nature of the rewards, we propose a region-based regularizer, aiming to preserve non-edited regions for high-reward samples while encouraging editing effects for low-reward samples. For evaluation, we annotate editing masks for GEdit-Bench and ImgEdit-Bench, introducing pixel-level similarity metrics to measure content consistency and editing quality. Applying CoCoEdit to Qwen-Image-Edit and FLUX-Kontext, we achieve not only competitive editing scores with state-of-the-art models, but also significantly better content consistency, measured by PSNR/SSIM metrics and human subjective ratings. --- ### REAL: Resolving Knowledge Conflicts in Knowledge-Intensive Visual Question Answering via Reasoning-Pivot Alignment Authors: Kai Ye, Xianwei Mao, Sheng Zhou, Zirui Shao, Ye Mo, Liangliang Liu, Haikuan Huang, Bin Li, Jiajun Bu Published: 2026-02-15 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.14065v2 Knowledge-intensive Visual Question Answering (KI-VQA) frequently suffers from severe knowledge conflicts caused by the inherent limitations of open-domain retrieval. However, existing paradigms face critical limitations due to the lack of generalizable conflict detection and intra-model constraint mechanisms to handle conflicting evidence. To address these challenges, we propose the REAL (Reasoning-Pivot Alignment) framework centered on the novel concept of the Reasoning-Pivot. Distinct from reasoning steps that prioritize internal self-derivation, a reasoning-pivot serves as an atomic unit (node or edge) in the reasoning chain that emphasizes knowledge linkage, and it typically relies on external evidence to complete the reasoning. Supported by our constructed REAL-VQA dataset, our approach integrates Reasoning-Pivot Aware SFT (RPA-SFT) to train a generalizable discriminator by aligning conflicts with pivot extraction, and employs Reasoning-Pivot Guided Decoding (RPGD), an intra-model decoding strategy that leverages these pivots for targeted conflict mitigation. Extensive experiments on diverse datasets demonstrate that REAL significantly enhances discrimination accuracy and achieves superior performance, validating our pivot-driven resolution paradigm. --- ### From Scarcity to Scale: A Release-Level Analysis of the Pashto Common Voice Dataset Authors: Jandad Jahani, Mursal Dawodi, Jawid Ahmad Baktash Published: 2026-02-15 Categories: cs.CL, cs.SD arXiv: https://arxiv.org/abs/2602.14062v1 Large, openly licensed speech datasets are essential for building automatic speech recognition (ASR) systems, yet many widely spoken languages remain underrepresented in public resources. Pashto, spoken by more than 60 million people, has historically lacked large-scale openly licensed speech data suitable for modern ASR development. This paper presents a release-level analysis of the Pashto component of the Mozilla Common Voice corpus, focusing on version 24.0 (December 2025) and contextualizing trends across major releases. We document rapid growth from 1.49 recorded hours in mid-2023 to 2,768.7 total hours in 2025, including 975.89 validated hours available for supervised ASR training. Beyond scale, we analyze validation throughput, contributor participation inequality, demographic metadata completeness, and sentence-level concentration in the validated subset. We find that participation is extremely concentrated (Gini = 0.941), age representation is strongly skewed toward young adults, and 41.97\% of clips lack self-reported gender labels, limiting subgroup auditing based on metadata. At the textual level, prompt reuse is moderate: 35.88\% of unique sentences account for 50\% of validated clips, suggesting that structural concentration is driven primarily by uneven contributor activity rather than dominance of a small prompt set. These results provide a quantitative audit of a rapidly scaling low-resource speech corpus and highlight practical priorities for improving dataset maturity, including expanded validation capacity and broader demographic participation. --- ### LM-Lexicon: Improving Definition Modeling via Harmonizing Semantic Experts Authors: Yang Liu, Jiaye Yang, Weikang Li, Jiahui Liang, Yang Li, Lingyong Yan Published: 2026-02-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.14060v1 We introduce LM-Lexicon, an innovative definition modeling approach that incorporates data clustering, semantic expert learning, and model merging using a sparse mixture-of-experts architecture. By decomposing the definition modeling task into specialized semantic domains, where small language models are trained as domain experts, LM-Lexicon achieves substantial improvements (+7% BLEU score compared with the prior state-of-the-art model) over existing methods on five widely used benchmarks. Empirically, we demonstrate that 1) the clustering strategy enables fine-grained expert specialization with nearly 10% improvement in definition quality; 2) the semantic-aware domain-level routing mechanism achieves higher expert efficacy (+1%) than conventional token-level routing; and 3) further performance gains can be obtained through test-time compute and semantic expert scaling. Our work advances definition modeling while providing insights into the development of efficient language models for semantic-intensive applications. --- ### LogitsCoder: Towards Efficient Chain-of-Thought Path Search via Logits Preference Decoding for Code Generation Authors: Jizheng Chen, Weiming Zhang, Xinyi Dai, Weiwen Liu, Kounianhua Du, Yasheng Wang, Ruiming Tang, Yong Yu, Weinan Zhang Published: 2026-02-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.14054v1 Code generation remains a challenging task that requires precise and structured reasoning. Existing Test Time Scaling (TTS) methods, including structured tree search, have made progress in exploring reasoning paths but still face two major challenges: (1) underthinking, where reasoning chains tend to be shallow and fail to capture the full complexity of problems; and (2) overthinking, where overly verbose reasoning leads to inefficiency and increased computational costs. To address these issues, we propose LogitsCoder, a novel framework that enhances chain-of-thought reasoning through lightweight, logit-level control mechanisms for code generation. LogitsCoder iteratively generates and refines reasoning steps by first steering token selection toward statistically preferred patterns via Logits Preference Decoding, then selecting and aggregating diverse reasoning paths using Logits Rank Based Path Selection and Thoughts Aggregation. This results in coherent and effective reasoning chains that balance depth and efficiency. Extensive experiments demonstrate that LogitsCoder produces more efficient and higher-quality reasoning chains, leading to superior code generation performance compared to baseline methods. --- ### Decentralized Federated Learning With Energy Harvesting Devices Authors: Kai Zhang, Xuanyu Cao, Khaled B. Letaief Published: 2026-02-15 Categories: cs.LG, eess.SP arXiv: https://arxiv.org/abs/2602.14051v1 Decentralized federated learning (DFL) enables edge devices to collaboratively train models through local training and fully decentralized device-to-device (D2D) model exchanges. However, these energy-intensive operations often rapidly deplete limited device batteries, reducing their operational lifetime and degrading the learning performance. To address this limitation, we apply energy harvesting technique to DFL systems, allowing edge devices to extract ambient energy and operate sustainably. We first derive the convergence bound for wireless DFL with energy harvesting, showing that the convergence is influenced by partial device participation and transmission packet drops, both of which further depend on the available energy supply. To accelerate convergence, we formulate a joint device scheduling and power control problem and model it as a multi-agent Markov decision process (MDP). Traditional MDP algorithms (e.g., value or policy iteration) require a centralized coordinator with access to all device states and exhibit exponential complexity in the number of devices, making them impractical for large-scale decentralized networks. To overcome these challenges, we propose a fully decentralized policy iteration algorithm that leverages only local state information from two-hop neighboring devices, thereby substantially reducing both communication overhead and computational complexity. We further provide a theoretical analysis showing that the proposed decentralized algorithm achieves asymptotic optimality. Finally, comprehensive numerical experiments on real-world datasets are conducted to validate the theoretical results and corroborate the effectiveness of the proposed algorithm. --- ### Position Encoding with Random Float Sampling Enhances Length Generalization of Transformers Authors: Atsushi Shimizu, Shohei Taniguchi, Yutaka Matsuo Published: 2026-02-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.14050v1 Length generalization is the ability of language models to maintain performance on inputs longer than those seen during pretraining. In this work, we introduce a simple yet powerful position encoding (PE) strategy, Random Float Sampling (RFS), that generalizes well to lengths unseen during pretraining or fine-tuning. In particular, instead of selecting position indices from a predefined discrete set, RFS uses randomly sampled continuous values, thereby avoiding out-of-distribution (OOD) issues on unseen lengths by exposing the model to diverse indices during training. Since assigning indices to tokens is a common and fundamental procedure in widely used PEs, the advantage of RFS can easily be incorporated into, for instance, the absolute sinusoidal encoding, RoPE, and ALiBi. Experiments corroborate its effectiveness by showing that RFS results in superior performance in length generalization tasks as well as zero-shot commonsense reasoning benchmarks. --- ### UniST-Pred: A Robust Unified Framework for Spatio-Temporal Traffic Forecasting in Transportation Networks Under Disruptions Authors: Yue Wang, Areg Karapetyan, Djellel Difallah, Samer Madanat Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14049v1 Spatio-temporal traffic forecasting is a core component of intelligent transportation systems, supporting various downstream tasks such as signal control and network-level traffic management. In real-world deployments, forecasting models must operate under structural and observational uncertainties, conditions that are rarely considered in model design. Recent approaches achieve strong short-term predictive performance by tightly coupling spatial and temporal modeling, often at the cost of increased complexity and limited modularity. In contrast, efficient time-series models capture long-range temporal dependencies without relying on explicit network structure. We propose UniST-Pred, a unified spatio-temporal forecasting framework that first decouples temporal modeling from spatial representation learning, then integrates both through adaptive representation-level fusion. To assess robustness of the proposed approach, we construct a dataset based on an agent-based, microscopic traffic simulator (MATSim) and evaluate UniST-Pred under severe network disconnection scenarios. Additionally, we benchmark UniST-Pred on standard traffic prediction datasets, demonstrating its competitive performance against existing well-established models despite a lightweight design. The results illustrate that UniST-Pred maintains strong predictive performance across both real-world and simulated datasets, while also yielding interpretable spatio-temporal representations under infrastructure disruptions. The source code and the generated dataset are available at https://anonymous.4open.science/r/UniST-Pred-EF27 --- ### ProAct: Harnessing Streaming Motion Generation and Agentic Reasoning for Real-Time Embodied Social Interaction Authors: Zeyi Zhang, Zixi Kang, Ruijie Zhao, Yusen Feng, Biao Jiang, Hanyu Ji, Libin Liu Published: 2026-02-15 Categories: cs.RO, cs.CV, cs.GR arXiv: https://arxiv.org/abs/2602.14048v2 Real-time embodied social interaction places two equally demanding requirements on an agent: continuously generating fluent multimodal interaction behavior, and proactively reasoning over accumulated dialogue and visual context to decide when to take initiative. These requirements must both be satisfied under a strict latency budget, making them difficult to meet simultaneously. We present ProAct, a dual-system framework that manages these time-critical requirements by integrating a low-latency Behavioral System for streaming multimodal interaction with a slower Cognitive System that performs long-horizon social reasoning and produces high-level proactive intentions. The Cognitive System incorporates an efficient memory mechanism and a user-motivation prediction module to reason over accumulated dialogue and visual context and determine when proactive intervention is appropriate. The Behavioral System further includes an intention-conditioned streaming flow-matching motion generator with a disentangled ControlNet branch, which translates deliberative intentions into continuous non-verbal behavior without disrupting interaction fluency. We deploy ProAct on a physical humanoid robot and validate the framework through comprehensive experiments, including real-world user studies, motion-generation benchmarks, and evaluation on ProActBench, a new, targeted benchmark for evaluating proactive trigger detection and restraint in embodied interaction. --- ### ProAct: A Dual-System Framework for Proactive Embodied Social Agents Authors: Zeyi Zhang, Zixi Kang, Ruijie Zhao, Yusen Feng, Biao Jiang, Libin Liu Published: 2026-02-15 Categories: cs.RO, cs.CV, cs.GR arXiv: https://arxiv.org/abs/2602.14048v1 Embodied social agents have recently advanced in generating synchronized speech and gestures. However, most interactive systems remain fundamentally reactive, responding only to current sensory inputs within a short temporal window. Proactive social behavior, in contrast, requires deliberation over accumulated context and intent inference, which conflicts with the strict latency budget of real-time interaction. We present \emph{ProAct}, a dual-system framework that reconciles this time-scale conflict by decoupling a low-latency \emph{Behavioral System} for streaming multimodal interaction from a slower \emph{Cognitive System} which performs long-horizon social reasoning and produces high-level proactive intentions. To translate deliberative intentions into continuous non-verbal behaviors without disrupting fluency, we introduce a streaming flow-matching model conditioned on intentions via ControlNet. This mechanism supports asynchronous intention injection, enabling seamless transitions between reactive and proactive gestures within a single motion stream. We deploy ProAct on a physical humanoid robot and evaluate both motion quality and interactive effectiveness. In real-world interaction user studies, participants and observers consistently prefer ProAct over reactive variants in perceived proactivity, social presence, and overall engagement, demonstrating the benefits of dual-system proactive control for embodied social interaction. --- ### Context Shapes LLMs Retrieval-Augmented Fact-Checking Effectiveness Authors: Pietro Bernardelle, Stefano Civelli, Kevin Roitero, Gianluca Demartini Published: 2026-02-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.14044v2 Large language models (LLMs) show strong reasoning abilities across diverse tasks, yet their performance on extended contexts remains inconsistent. While prior research has emphasized mid-context degradation in question answering, this study examines the impact of context in LLM-based fact verification. Using three datasets (HOVER, FEVEROUS, and ClimateFEVER) and five open-source models accross different parameters sizes (7B, 32B and 70B parameters) and model families (Llama-3.1, Qwen2.5 and Qwen3), we evaluate both parametric factual knowledge and the impact of evidence placement across varying context lengths. We find that LLMs exhibit non-trivial parametric knowledge of factual claims and that their verification accuracy generally declines as context length increases. Similarly to what has been shown in previous works, in-context evidence placement plays a critical role with accuracy being consistently higher when relevant evidence appears near the beginning or end of the prompt and lower when placed mid-context. These results underscore the importance of prompt structure in retrieval-augmented fact-checking systems. --- ### Beyond Static Snapshots: Dynamic Modeling and Forecasting of Group-Level Value Evolution with Large Language Models Authors: Qiankun Pi, Guixin Su, Jinliang Li, Mayi Xu, Xin Miao, Jiawei Jiang, Ming Zhong, Tieyun Qian Published: 2026-02-15 Categories: cs.SI, cs.AI arXiv: https://arxiv.org/abs/2602.14043v1 Social simulation is critical for mining complex social dynamics and supporting data-driven decision making. LLM-based methods have emerged as powerful tools for this task by leveraging human-like social questionnaire responses to model group behaviors. Existing LLM-based approaches predominantly focus on group-level values at discrete time points, treating them as static snapshots rather than dynamic processes. However, group-level values are not fixed but shaped by long-term social changes. Modeling their dynamics is thus crucial for accurate social evolution prediction--a key challenge in both data mining and social science. This problem remains underexplored due to limited longitudinal data, group heterogeneity, and intricate historical event impacts. To bridge this gap, we propose a novel framework for group-level dynamic social simulation by integrating historical value trajectories into LLM-based human response modeling. We select China and the U.S. as representative contexts, conducting stratified simulations across four core sociodemographic dimensions (gender, age, education, income). Using the World Values Survey, we construct a multi-wave, group-level longitudinal dataset to capture historical value evolution, and then propose the first event-based prediction method for this task, unifying social events, current value states, and group attributes into a single framework. Evaluations across five LLM families show substantial gains: a maximum 30.88\% improvement on seen questions and 33.97\% on unseen questions over the Vanilla baseline. We further find notable cross-group heterogeneity: U.S. groups are more volatile than Chinese groups, and younger groups in both countries are more sensitive to external changes. These findings advance LLM-based social simulation and provide new insights for social scientists to understand and predict social value changes. --- ### Restoration Adaptation for Semantic Segmentation on Low Quality Images Authors: Kai Guan, Rongyuan Wu, Shuai Li, Wentao Zhu, Wenjun Zeng, Lei Zhang Published: 2026-02-15 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.14042v1 In real-world scenarios, the performance of semantic segmentation often deteriorates when processing low-quality (LQ) images, which may lack clear semantic structures and high-frequency details. Although image restoration techniques offer a promising direction for enhancing degraded visual content, conventional real-world image restoration (Real-IR) models primarily focus on pixel-level fidelity and often fail to recover task-relevant semantic cues, limiting their effectiveness when directly applied to downstream vision tasks. Conversely, existing segmentation models trained on high-quality data lack robustness under real-world degradations. In this paper, we propose Restoration Adaptation for Semantic Segmentation (RASS), which effectively integrates semantic image restoration into the segmentation process, enabling high-quality semantic segmentation on the LQ images directly. Specifically, we first propose a Semantic-Constrained Restoration (SCR) model, which injects segmentation priors into the restoration model by aligning its cross-attention maps with segmentation masks, encouraging semantically faithful image reconstruction. Then, RASS transfers semantic restoration knowledge into segmentation through LoRA-based module merging and task-specific fine-tuning, thereby enhancing the model's robustness to LQ images. To validate the effectiveness of our framework, we construct a real-world LQ image segmentation dataset with high-quality annotations, and conduct extensive experiments on both synthetic and real-world LQ benchmarks. The results show that SCR and RASS significantly outperform state-of-the-art methods in segmentation and restoration tasks. Code, models, and datasets will be available at https://github.com/Ka1Guan/RASS.git. --- ### BitDance: Scaling Autoregressive Generative Models with Binary Tokens Authors: Yuang Ai, Jiaming Han, Shaobin Zhuang, Weijia Mao, Xuefeng Hu, Ziyan Yang, Zhenheng Yang, Yali Wang, Huaibo Huang, Xiangyu Yue, Hao Chen Published: 2026-02-15 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.14041v2 We present BitDance, a scalable autoregressive (AR) image generator that predicts binary visual tokens instead of codebook indices. With high-entropy binary latents, BitDance lets each token represent up to $2^{256}$ states, yielding a compact yet highly expressive discrete representation. Sampling from such a huge token space is difficult with standard classification. To resolve this, BitDance uses a binary diffusion head: instead of predicting an index with softmax, it employs continuous-space diffusion to generate the binary tokens. Furthermore, we propose next-patch diffusion, a new decoding method that predicts multiple tokens in parallel with high accuracy, greatly speeding up inference. On ImageNet 256x256, BitDance achieves an FID of 1.24, the best among AR models. With next-patch diffusion, BitDance beats state-of-the-art parallel AR models that use 1.4B parameters, while using 5.4x fewer parameters (260M) and achieving 8.7x speedup. For text-to-image generation, BitDance trains on large-scale multimodal tokens and generates high-resolution, photorealistic images efficiently, showing strong performance and favorable scaling. When generating 1024x1024 images, BitDance achieves a speedup of over 30x compared to prior AR models. We release code and models to facilitate further research on AR foundation models. Code and models are available at: https://github.com/shallowdream204/BitDance. --- ### Explainability-Inspired Layer-Wise Pruning of Deep Neural Networks for Efficient Object Detection Authors: Abhinav Shukla, Nachiket Tapas Published: 2026-02-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.14040v1 Deep neural networks (DNNs) have achieved remarkable success in object detection tasks, but their increasing complexity poses significant challenges for deployment on resource-constrained platforms. While model compression techniques such as pruning have emerged as essential tools, traditional magnitude-based pruning methods do not necessarily align with the true functional contribution of network components to task-specific performance. In this work, we present an explainability-inspired, layer-wise pruning framework tailored for efficient object detection. Our approach leverages a SHAP-inspired gradient--activation attribution to estimate layer importance, providing a data-driven proxy for functional contribution rather than relying solely on static weight magnitudes. We conduct comprehensive experiments across diverse object detection architectures, including ResNet-50, MobileNetV2, ShuffleNetV2, Faster R-CNN, RetinaNet, and YOLOv8, evaluating performance on the Microsoft COCO 2017 validation set. The results show that the proposed attribution-inspired pruning consistently identifies different layers as least important compared to L1-norm-based methods, leading to improved accuracy--efficiency trade-offs. Notably, for ShuffleNetV2, our method yields a 10\% empirical increase in inference speed, whereas L1-pruning degrades performance by 13.7\%. For RetinaNet, the proposed approach preserves the baseline mAP (0.151) with negligible impact on inference speed, while L1-pruning incurs a 1.3\% mAP drop for a 6.2\% speed increase. These findings highlight the importance of data-driven layer importance assessment and demonstrate that explainability-inspired compression offers a principled direction for deploying deep neural networks on edge and resource-constrained platforms while preserving both performance and interpretability. --- ### Geometry-Preserving Aggregation for Mixture-of-Experts Embedding Models Authors: Sajjad Kachuee, Mohammad Sharifkhani Published: 2026-02-15 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.14039v1 Mixture-of-Experts (MoE) embedding models combine expert outputs using weighted linear summation, implicitly assuming a linear subspace structure in the embedding space. This assumption is shown to be inconsistent with the geometry of expert representations. Geometric analysis of a modern MoE embedding model reveals that expert outputs lie on a shared hyperspherical manifold characterized by tightly concentrated norms and substantial angular separation. Under this geometry, linear aggregation induces inward collapse toward the manifold interior, distorting vector magnitude and direction and reducing embedding comparability. To address this inconsistency, Spherical Barycentric Aggregation (SBA) is introduced as a geometry-preserving aggregation operator that separates radial and angular components to maintain hyperspherical structure while remaining fully compatible with existing routing mechanisms. Experiments on selected tasks from the Massive Text Embedding Benchmark (MTEB), including semantic similarity, clustering, and duplicate question detection, demonstrate consistent performance improvements with identical training cost and full stability. Additional geometric analyses confirm that SBA prevents aggregation-induced collapse and preserves hyperspherical consistency, highlighting the importance of geometry-aware aggregation in MoE embedding architectures. --- ### Choosing How to Remember: Adaptive Memory Structures for LLM Agents Authors: Mingfei Lu, Mengjia Wu, Feng Liu, Jiawei Xu, Weikai Li, Haoyang Wang, Zhengdong Hu, Ying Ding, Yizhou Sun, Jie Lu, Yi Zhang Published: 2026-02-15 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.14038v1 Memory is critical for enabling large language model (LLM) based agents to maintain coherent behavior over long-horizon interactions. However, existing agent memory systems suffer from two key gaps: they rely on a one-size-fits-all memory structure and do not model memory structure selection as a context-adaptive decision, limiting their ability to handle heterogeneous interaction patterns and resulting in suboptimal performance. We propose a unified framework, FluxMem, that enables adaptive memory organization for LLM agents. Our framework equips agents with multiple complementary memory structures. It explicitly learns to select among these structures based on interaction-level features, using offline supervision derived from downstream response quality and memory utilization. To support robust long-horizon memory evolution, we further introduce a three-level memory hierarchy and a Beta Mixture Model-based probabilistic gate for distribution-aware memory fusion, replacing brittle similarity thresholds. Experiments on two long-horizon benchmarks, PERSONAMEM and LoCoMo, demonstrate that our method achieves average improvements of 9.18% and 6.14%. --- ### FloCA: Towards Faithful and Logically Consistent Flowchart Reasoning Authors: Jinzi Zou, Bolin Wang, Shuo Zhang, Nuo Xu, Junzhou Zhao Published: 2026-02-15 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.14035v2 Flowchart-oriented dialogue (FOD) systems aim to guide users through multi-turn decision-making or operational procedures by following a domain-specific flowchart to achieve a task goal. In this work, we formalize flowchart reasoning in FOD as grounding user input to flowchart nodes at each dialogue turn while ensuring node transition is consistent with the correct flowchart path. Despite recent advances of LLMs in task-oriented dialogue systems, adapting them to FOD still faces two limitations: (1) LLMs lack an explicit mechanism to represent and reason over flowchart topology, and (2) they are prone to hallucinations, leading to unfaithful flowchart reasoning. To address these limitations, we propose FloCA, a zero-shot flowchart-oriented conversational agent. FloCA uses an LLM for intent understanding and response generation, while delegating flowchart reasoning to an external tool that performs topology-constrained graph execution, ensuring faithful and logically consistent node transitions across dialogue turns. We further introduce an evaluation framework with an LLM-based user simulator and five new metrics covering reasoning accuracy and interaction efficiency. Extensive experiments on FLODIAL and PFDial datasets highlight the bottlenecks of existing LLM/VLM-based methods and demonstrate the superiority of FloCA. The code and dataset are publicly available at https://github.com/Jinzi-Zou/FloCA-flowchart-reasoning. --- ### FloCA: Towards Faithful and Logically Consistent Flowchart Reasoning Authors: Jinzi Zou, Bolin Wang, Liang Li, Shuo Zhang, Nuo Xu, Junzhou Zhao Published: 2026-02-15 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.14035v1 Flowchart-oriented dialogue (FOD) systems aim to guide users through multi-turn decision-making or operational procedures by following a domain-specific flowchart to achieve a task goal. In this work, we formalize flowchart reasoning in FOD as grounding user input to flowchart nodes at each dialogue turn while ensuring node transition is consistent with the correct flowchart path. Despite recent advances of LLMs in task-oriented dialogue systems, adapting them to FOD still faces two limitations: (1) LLMs lack an explicit mechanism to represent and reason over flowchart topology, and (2) they are prone to hallucinations, leading to unfaithful flowchart reasoning. To address these limitations, we propose FloCA, a zero-shot flowchart-oriented conversational agent. FloCA uses an LLM for intent understanding and response generation while delegating flowchart reasoning to an external tool that performs topology-constrained graph execution, ensuring faithful and logically consistent node transitions across dialogue turns. We further introduce an evaluation framework with an LLM-based user simulator and five new metrics covering reasoning accuracy and interaction efficiency. Extensive experiments on FLODIAL and PFDial datasets highlight the bottlenecks of existing LLM-based methods and demonstrate the superiority of FloCA. Our codes are available at https://github.com/Jinzi-Zou/FloCA-flowchart-reasoning. --- ### BRAIN: Bayesian Reasoning via Active Inference for Agentic and Embodied Intelligence in Mobile Networks Authors: Osman Tugay Basaran, Martin Maier, Falko Dressler Published: 2026-02-15 Categories: cs.IT, cs.AI arXiv: https://arxiv.org/abs/2602.14033v1 Future sixth-generation (6G) mobile networks will demand artificial intelligence (AI) agents that are not only autonomous and efficient, but also capable of real-time adaptation in dynamic environments and transparent in their decisionmaking. However, prevailing agentic AI approaches in networking, exhibit significant shortcomings in this regard. Conventional deep reinforcement learning (DRL)-based agents lack explainability and often suffer from brittle adaptation, including catastrophic forgetting of past knowledge under non-stationary conditions. In this paper, we propose an alternative solution for these challenges: Bayesian reasoning via Active Inference (BRAIN) agent. BRAIN harnesses a deep generative model of the network environment and minimizes variational free energy to unify perception and action in a single closed-loop paradigm. We implement BRAIN as O-RAN eXtended application (xApp) on GPU-accelerated testbed and demonstrate its advantages over standard DRL baselines. In our experiments, BRAIN exhibits (i) robust causal reasoning for dynamic radio resource allocation, maintaining slice-specific quality of service (QoS) targets (throughput, latency, reliability) under varying traffic loads, (ii) superior adaptability with up to 28.3% higher robustness to sudden traffic shifts versus benchmarks (achieved without any retraining), and (iii) real-time interpretability of its decisions through human-interpretable belief state diagnostics. --- ### MC$^2$Mark: Distortion-Free Multi-Bit Watermarking for Long Messages Authors: Xuehao Cui, Ruibo Chen, Yihan Wu, Heng Huang Published: 2026-02-15 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2602.14030v1 Large language models now produce text indistinguishable from human writing, which increases the need for reliable provenance tracing. Multi-bit watermarking can embed identifiers into generated text, but existing methods struggle to keep both text quality and watermark strength while carrying long messages. We propose MC$^2$Mark, a distortion-free multi-bit watermarking framework designed for reliable embedding and decoding of long messages. Our key technical idea is Multi-Channel Colored Reweighting, which encodes bits through structured token reweighting while keeping the token distribution unbiased, together with Multi-Layer Sequential Reweighting to strengthen the watermark signal and an evidence-accumulation detector for message recovery. Experiments show that MC$^2$Mark improves detectability and robustness over prior multi-bit watermarking methods while preserving generation quality, achieving near-perfect accuracy for short messages and exceeding the second-best method by nearly 30% for long messages. --- ### Why Self-Training Helps and Hurts: Denoising vs. Signal Forgetting Authors: Mingqi Wu, Archer Y. Yang, Qiang Sun Published: 2026-02-15 Categories: stat.ML, cs.LG, math.ST arXiv: https://arxiv.org/abs/2602.14029v1 Iterative self-training (self-distillation) repeatedly refits a model on pseudo-labels generated by its own predictions. We study this procedure in overparameterized linear regression: an initial estimator is trained on noisy labels, and each subsequent iterate is trained on fresh covariates with noiseless pseudo-labels from the previous model. In the high-dimensional regime, we derive deterministic-equivalent recursions for the prediction risk and effective noise across iterations, and prove that the empirical quantities concentrate sharply around these limits. The recursion separates two competing forces: a systematic component that grows with iteration due to progressive signal forgetting, and a stochastic component that decays due to denoising via repeated data-dependent projections. Their interaction yields a $U$-shaped test-risk curve and an optimal early-stopping time. In spiked covariance models, iteration further acts as an iteration-dependent spectral filter that preserves strong eigendirections while suppressing weaker ones, inducing an implicit form of soft feature selection distinct from ridge regression. Finally, we propose an iterated generalized cross-validation criterion and prove its uniform consistency for estimating the risk along the self-training trajectory, enabling fully data-driven selection of the stopping time and regularization. Experiments on synthetic covariances validate the theory and illustrate the predicted denoising-forgetting trade-off. --- ### GRRM: Group Relative Reward Modeling for Machine Translation Authors: Sen Yang, Shanbo Cheng, Lu Xu, Jianbing Zhang, Shujian Huang Published: 2026-02-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.14028v2 While Group Relative Policy Optimization (GRPO) offers a powerful framework for LLM post-training, its effectiveness in open-ended domains like Machine Translation hinges on accurate intra-group ranking. We identify that standard Pointwise Quality Metrics (PQM) fall short in this context: candidates are evaluated in isolation, so the comparative context is missing for distinguishing fine-grained linguistic nuances. To address this, we introduce the Group Quality Metric (GQM) paradigm and its instantiation, the Group Relative Reward Model (GRRM). Unlike traditional independent scorers, GRRM jointly processes the entire candidate group, leveraging comparative analysis to rigorously resolve relative quality and adaptive granularity. Empirical evaluations confirm that GRRM achieves competitive ranking accuracy among all baselines; integrating GRRM into the GRPO training not only improves general translation quality but also unlocks reasoning capabilities comparable to state-of-the-art reasoning models. We release codes, datasets, and model checkpoints at https://github.com/NJUNLP/GRRM. --- ### GRRM: Group Relative Reward Modeling for Machine Translation Authors: Sen Yang, Shanbo Cheng, Lu Xu, Jianbing Zhang, Shujian Huang Published: 2026-02-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.14028v1 While Group Relative Policy Optimization (GRPO) offers a powerful framework for LLM post-training, its effectiveness in open-ended domains like Machine Translation hinges on accurate intra-group ranking. We identify that standard Scalar Quality Metrics (SQM) fall short in this context; by evaluating candidates in isolation, they lack the comparative context necessary to distinguish fine-grained linguistic nuances. To address this, we introduce the Group Quality Metric (GQM) paradigm and instantiate it via the Group Relative Reward Model (GRRM). Unlike traditional independent scorers, GRRM processes the entire candidate group jointly, leveraging comparative analysis to rigorously resolve relative quality and adaptive granularity. Empirical evaluations confirm that GRRM achieves competitive ranking accuracy among all baselines. Building on this foundation, we integrate GRRM into the GRPO training loop to optimize the translation policy. Experimental results demonstrate that our framework not only improves general translation quality but also unlocks reasoning capabilities comparable to state-of-the-art reasoning models. We release codes, datasets, and model checkpoints at https://github.com/NJUNLP/GRRM. --- ### Train Short, Inference Long: Training-free Horizon Extension for Autoregressive Video Generation Authors: Jia Li, Xiaomeng Fu, Xurui Peng, Weifeng Chen, Youwei Zheng, Tianyu Zhao, Jiexi Wang, Fangmin Chen, Xing Wang, Hayden Kwok-Hay So Published: 2026-02-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.14027v3 Autoregressive video diffusion models have emerged as a scalable paradigm for long video generation. However, they often suffer from severe extrapolation failure, where rapid error accumulation leads to significant temporal degradation when extending beyond training horizons. We identify that this failure primarily stems from the spectral bias of 3D positional embeddings and the lack of dynamic priors in noise sampling. To address these issues, we propose FLEX (Frequency-aware Length EXtension), a training-free inference-time framework that bridges the gap between short-term training and long-term inference. FLEX introduces Frequency-aware RoPE Modulation to adaptively interpolate under-trained low-frequency components while extrapolating high-frequency ones to preserve multi-scale temporal discriminability. This is integrated with Antiphase Noise Sampling (ANS) to inject high-frequency dynamic priors and Inference-only Attention Sink to anchor global structure. Extensive evaluations on VBench demonstrate that FLEX significantly outperforms state-of-the-art models at 6x extrapolation (30s duration) and matches the performance of long-video fine-tuned baselines at 12x scale (60s duration). As a plug-and-play augmentation, FLEX seamlessly integrates into existing inference pipelines for horizon extension. It effectively pushes the generation limits of models such as LongLive, supporting consistent and dynamic video synthesis at a 4-minute scale. Project page is available at https://ga-lee.github.io/FLEX_demo. --- ### EIDOS: Latent-Space Predictive Learning for Time Series Foundation Models Authors: Xinxing Zhou, Qingren Yao, Yiji Zhao, Chenghao Liu, Flora Salim, Xiaojie Yuan, Yanlong Wen, Ming Jin Published: 2026-02-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.14024v1 Most time series foundation models are pretrained by directly predicting future observations, which often yields weakly structured latent representations that capture surface noise rather than coherent and predictable temporal dynamics. In this work, we introduce EIDOS, a foundation model family that shifts pretraining from future value prediction to latent-space predictive learning. We train a causal Transformer to predict the evolution of latent representations, encouraging the emergence of structured and temporally coherent latent states. To ensure stable targets for latent-space learning, we design a lightweight aggregation branch to construct target representations. EIDOS is optimized via a joint objective that integrates latent-space alignment, observational grounding to anchor representations to the input signal, and direct forecasting supervision. On the GIFT-Eval benchmark, EIDOS mitigates structural fragmentation in the representation space and achieves state-of-the-art performance. These results demonstrate that constraining models to learn predictable latent dynamics is a principled step toward more robust and reliable time series foundation models. --- ### Flow4R: Unifying 4D Reconstruction and Tracking with Scene Flow Authors: Shenhan Qian, Ganlin Zhang, Shangzhe Wu, Daniel Cremers Published: 2026-02-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.14021v2 Reconstructing and tracking dynamic 3D scenes is a fundamental challenge in computer vision. Existing methods typically decouple geometry from motion: static multi-view reconstruction systems assume a rigid world, whereas dynamic tracking frameworks rely on explicit ego-motion estimation or separate object motion models. In this work, we propose Flow4R, a unified framework that treats relative scene flow as the central representation linking 3D structure, camera ego-motion, and dynamic object motion. Given a two-view input, Flow4R employs a shared Vision Transformer to predict a compact, pixel-aligned property set comprising 3D point positions, scene flow, pose weights, and confidence maps. This flow-centric formulation allows local geometry and bidirectional motion to be jointly inferred in a single feedforward pass, eliminating the need for explicit pose regression heads or complex bundle adjustment. By training jointly on static and dynamic datasets, Flow4R achieves state-of-the-art performance on 4D reconstruction and tracking benchmarks, demonstrating the power of the flow-centric formulation for spatiotemporal scene understanding. --- ### Computable Bernstein Certificates for Cross-Fitted Clipped Covariance Estimation Authors: Even He, Zaizai Yan Published: 2026-02-15 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2602.14020v1 We study operator-norm covariance estimation from heavy-tailed samples that may include a small fraction of arbitrary outliers. A simple and widely used safeguard is \emph{Euclidean norm clipping}, but its accuracy depends critically on an unknown clipping level. We propose a cross-fitted clipped covariance estimator equipped with \emph{fully computable} Bernstein-type deviation certificates, enabling principled data-driven tuning via a selector (\emph{MinUpper}) that balances certified stochastic error and a robust hold-out proxy for clipping bias. The resulting procedure adapts to intrinsic complexity measures such as effective rank under mild tail regularity and retains meaningful guarantees under only finite fourth moments. Experiments on contaminated spiked-covariance benchmarks illustrate stable performance and competitive accuracy across regimes. --- ### S2SServiceBench: A Multimodal Benchmark for Last-Mile S2S Climate Services Authors: Chenyue Li, Wen Deng, Zhuotao Sun, Mengxi Jin, Hanzhe Cui, Han Li, Shentong Li, Man Kit Yu, Ming Long Lai, Yuhao Yang, Mengqian Lu, Binhang Yuan Published: 2026-02-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.14017v1 Subseasonal-to-seasonal (S2S) forecasts play an essential role in providing a decision-critical weeks-to-months planning window for climate resilience and sustainability, yet a growing bottleneck is the last-mile gap: translating scientific forecasts into trusted, actionable climate services, requiring reliable multimodal understanding and decision-facing reasoning under uncertainty. Meanwhile, multimodal large language models (MLLMs) and corresponding agentic paradigms have made rapid progress in supporting various workflows, but it remains unclear whether they can reliably generate decision-making deliverables from operational service products (e.g., actionable signal comprehension, decision-making handoff, and decision analysis & planning) under uncertainty. We introduce S2SServiceBench, a multimodal benchmark for last-mile S2S climate services curated from an operational climate-service system to evaluate this capability. S2SServiceBenchcovers 10 service products with about 150+ expert-selected cases in total, spanning six application domains - Agriculture, Disasters, Energy, Finance, Health, and Shipping. Each case is instantiated at three service levels, yielding around 500 tasks and 1,000+ evaluation items across climate resilience and sustainability applications. Using S2SServiceBench, we benchmark state-of-the-art MLLMs and agents, and analyze performance across products and service levels, revealing persistent challenges in S2S service plot understanding and reasoning - namely, actionable signal comprehension, operationalizing uncertainty into executable handoffs, and stable, evidence-grounded analysis and planning for dynamic hazards-while offering actionable guidance for building future climate-service agents. --- ### From SFT to RL: Demystifying the Post-Training Pipeline for LLM-based Vulnerability Detection Authors: Youpeng Li, Fuxun Yu, Xinda Wang Published: 2026-02-15 Categories: cs.CR, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2602.14012v2 The integration of LLMs into vulnerability detection (VD) has shifted the field toward more interpretable and context-aware analysis. While post-training techniques have shown promise in general coding tasks, their systematic application to VD remains underexplored. In this paper, we present the first comprehensive investigation into the post-training pipeline for LLM-based VD, demonstrating that on-policy RL with GRPO consistently outperforms SFT, off-policy preference optimization methods, and specialized VD LLMs. Our study further reveals VD-specific post-training guidelines and insights beyond common practices: (1) For data curation, contrary to the widespread use of rationalization-based supervision in prior VD work, SFT based on rejection sampling proves more effective, as rationalization can introduce hallucinations; in RL training, the inherently skewed difficulty distribution of vulnerabilities leads difficulty-aware data filtering to drastically reduce data coverage, causing non-negligible performance loss, and undermines curriculum learning, while pair-based data scheduling can partially mitigate this. (2) For stage interactions, unlike preference optimization typically applied to lightly trained SFT models, increasing SFT epochs consistently benefits off-policy preference optimization in VD tasks; however, excessive SFT suppresses self-exploration in on-policy RL, limiting its gains. (3) For reward mechanisms, naively treating vulnerability classification correctness as reward signals leads to reward hacking, whereas fine-grained root-cause judgments provide more reliable credit assignment; specification-based rewards further improve efficiency at the cost of additional design and generation effort. (4) For evaluation protocols, LLM-as-a-Judge based on root-cause analysis offers a more robust alternative, albeit with variability across judge models. --- ### KoopGen: Koopman Generator Networks for Representing and Predicting Dynamical Systems with Continuous Spectra Authors: Liangyu Su, Jun Shu, Rui Liu, Deyu Meng, Zongben Xu Published: 2026-02-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.14011v1 Representing and predicting high-dimensional and spatiotemporally chaotic dynamical systems remains a fundamental challenge in dynamical systems and machine learning. Although data-driven models can achieve accurate short-term forecasts, they often lack stability, interpretability, and scalability in regimes dominated by broadband or continuous spectra. Koopman-based approaches provide a principled linear perspective on nonlinear dynamics, but existing methods rely on restrictive finite-dimensional assumptions or explicit spectral parameterizations that degrade in high-dimensional settings. Against these issues, we introduce KoopGen, a generator-based neural Koopman framework that models dynamics through a structured, state-dependent representation of Koopman generators. By exploiting the intrinsic Cartesian decomposition into skew-adjoint and self-adjoint components, KoopGen separates conservative transport from irreversible dissipation while enforcing exact operator-theoretic constraints during learning. Across systems ranging from nonlinear oscillators to high-dimensional chaotic and spatiotemporal dynamics, KoopGen improves prediction accuracy and stability, while clarifying which components of continuous-spectrum dynamics admit interpretable and learnable representations. --- ### A Deployment-Friendly Foundational Framework for Efficient Computational Pathology Authors: Yu Cai, Cheng Jin, Zhengyu Zhang, Jiabo Ma, Fengtao Zhou, Yingxue Xu, Zhengrui Guo, Yihui Wang, Zhengyu Zhang, Ling Liang, Yonghao Tan, Pingcheng Dong et al. Published: 2026-02-15 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.14010v2 Pathology foundation models (PFMs) generalize well across computational pathology tasks but remain costly for gigapixel whole-slide image analysis. Here, we present LitePath, a deployment-friendly framework that addresses model over-parameterization and patch-level redundancy. LitePath combines LiteFM, a compact model distilled from Virchow2, H-Optimus-1 and UNI2 using 190 million patches, with the Adaptive Patch Selector for task-specific patch selection. Compared with Virchow2, LitePath uses 28x fewer parameters and 403.5x fewer FLOPs. On an NVIDIA Jetson Orin Nano Super, it processes 208 slides per hour, 104.5x faster than Virchow2, and consumes 0.36 kWh per 3,000 slides, 171x less energy than Virchow2 on an RTX 3090 GPU. We evaluated LitePath on 45 multicenter cohorts across four organs and 33 tasks, comprising 37 classification and 8 survival cohorts, including 33 internal, 10 external and 2 prospective cohorts, with 17,837 slides from 9,977 patients disjoint from the pretraining data. Among 22 PFMs, LitePath achieved the best average rank (6.56 vs. 6.58 for Virchow2), retained 99.71% of Virchow2's Macro-AUC across classification cohorts, and improved mean C-index by 2.14 percentage points across survival cohorts (71.91% vs. 69.77%). We further introduce the Deployability Score (D-Score), a weighted geometric mean of normalized task performance and FLOPs. LitePath achieved the highest D-Score (0.8455), outperforming H0-mini (0.7723) and Virchow2 (0.7297). In a randomized paired crossover study of four pathologists and 120 cases, LitePath increased diagnostic accuracy by 4.1-15.8 percentage points and reduced diagnostic time by 10.9-14.2%. These results demonstrate rapid, cost-effective and energy-efficient pathology image analysis on accessible hardware while maintaining competitive performance. ---