# Research Papers — 2026-01 · bundle page 23 of 189 > Full-text markdown of papers #1101–1150 of 9403 published in 2026-01, newest first. AI-readable mirrors of the full 9403-paper 2026-01 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-01/page/23 [← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/22.md) [Next bundle →](https://aipod.dev/papers/archive/2026-01/page/24.md) ### Mechanistic Data Attribution: Tracing the Training Origins of Interpretable LLM Units Authors: Jianhui Chen, Yuzhang Luo, Liangming Pan Published: 2026-01-29 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.21996v2 While Mechanistic Interpretability has identified interpretable circuits in LLMs, their causal origins in training data remain elusive. We introduce Mechanistic Data Attribution (MDA), a scalable framework that employs Influence Functions to trace interpretable units back to specific training samples. Through extensive experiments on the Pythia family, we causally validate that targeted intervention--removing or augmenting a small fraction of high-influence samples--significantly modulates the emergence of interpretable heads, whereas random interventions show no effect. Our analysis reveals that repetitive structural data (e.g., LaTeX, XML) acts as a mechanistic catalyst. Furthermore, we observe that interventions targeting induction head formation induce a concurrent change in the model's in-context learning (ICL) capability. This provides direct causal evidence for the long-standing hypothesis regarding the functional link between induction heads and ICL. Finally, we propose a mechanistic data augmentation pipeline that consistently accelerates circuit convergence across model scales, providing a principled methodology for steering the developmental trajectories of LLMs. --- ### Liquid Interfaces: A Dynamic Ontology for the Interoperability of Autonomous Systems Authors: Dhiogo de Sá, Carlos Schmiedel, Carlos Pereira Lopes Published: 2026-01-29 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2601.21993v1 Contemporary software architectures struggle to support autonomous agents whose reasoning is adaptive, probabilistic, and context-dependent, while system integration remains dominated by static interfaces and deterministic contracts. This paper introduces Liquid Interfaces, a coordination paradigm in which interfaces are not persistent technical artifacts, but ephemeral relational events that emerge through intention articulation and semantic negotiation at runtime.We formalize this model and present the Liquid Interface Protocol (LIP),which governs intention-driven interaction, negotiated execution, and enforce ephemerality under semantic uncertainty. We further discuss the governance implications of this approach and describe a reference architecture that demonstrates practical feasibility. Liquid Interfaces provide a principled foundation for adaptive coordination in agent-based systems --- ### Geometry of Drifting MDPs with Path-Integral Stability Certificates Authors: Zuyuan Zhang, Mahdi Imani, Tian Lan Published: 2026-01-29 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.21991v1 Real-world reinforcement learning is often \emph{nonstationary}: rewards and dynamics drift, accelerate, oscillate, and trigger abrupt switches in the optimal action. Existing theory often represents nonstationarity with coarse-scale models that measure \emph{how much} the environment changes, not \emph{how} it changes locally -- even though acceleration and near-ties drive tracking error and policy chattering. We take a geometric view of nonstationary discounted Markov Decision Processes (MDPs) by modeling the environment as a differentiable homotopy path and tracking the induced motion of the optimal Bellman fixed point. This yields a length-curvature-kink signature of intrinsic complexity: cumulative drift, acceleration/oscillation, and action-gap-induced nonsmoothness. We prove a solver-agnostic path-integral stability bound and derive gap-safe feasible regions that certify local stability away from switch regimes. Building on these results, we introduce \textit{Homotopy-Tracking RL (HT-RL)} and \textit{HT-MCTS}, lightweight wrappers that estimate replay-based proxies of length, curvature, and near-tie proximity online and adapt learning or planning intensity accordingly. Experiments show improved tracking and dynamic regret over matched static baselines, with the largest gains in oscillatory and switch-prone regimes. --- ### Batched First-Order Methods for Parallel LP Solving in MIP Authors: Nicolas Blin, Stefano Gualandi, Christopher Maes, Andrea Lodi, Bartolomeo Stellato Published: 2026-01-29 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2601.21990v1 We present a batched first-order method for solving multiple linear programs in parallel on GPUs. Our approach extends the primal-dual hybrid gradient algorithm to efficiently solve batches of related linear programming problems that arise in mixed-integer programming techniques such as strong branching and bound tightening. By leveraging matrix-matrix operations instead of repeated matrix-vector operations, we obtain significant computational advantages on GPU architectures. We demonstrate the effectiveness of our approach on various case studies and identify the problem sizes where first-order methods outperform traditional simplex-based solvers depending on the computational environment one can use. This is a significant step for the design and development of integer programming algorithms tightly exploiting GPU capabilities where we argue that some specific operations should be allocated to GPUs and performed in full instead of using light-weight heuristic approaches on CPUs. --- ### Generalized Information Gathering Under Dynamics Uncertainty Authors: Fernando Palafox, Jingqi Li, Jesse Milzman, David Fridovich-Keil Published: 2026-01-29 Categories: cs.LG, cs.AI, cs.MA, cs.RO, eess.SY arXiv: https://arxiv.org/abs/2601.21988v1 An agent operating in an unknown dynamical system must learn its dynamics from observations. Active information gathering accelerates this learning, but existing methods derive bespoke costs for specific modeling choices: dynamics models, belief update procedures, observation models, and planners. We present a unifying framework that decouples these choices from the information-gathering cost by explicitly exposing the causal dependencies between parameters, beliefs, and controls. Using this framework, we derive a general information-gathering cost based on Massey's directed information that assumes only Markov dynamics with additive noise and is otherwise agnostic to modeling choices. We prove that the mutual information cost used in existing literature is a special case of our cost. Then, we leverage our framework to establish an explicit connection between the mutual information cost and information gain in linearized Bayesian estimation, thereby providing theoretical justification for mutual information-based active learning approaches. Finally, we illustrate the practical utility of our framework through experiments spanning linear, nonlinear, and multi-agent systems. --- ### Elign: Equivariant Diffusion Model Alignment from Foundational Machine Learning Force Fields Authors: Yunyang Li, Lin Huang, Luojia Xia, Wenhe Zhang, Mark Gerstein Published: 2026-01-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.21985v1 Generative models for 3D molecular conformations must respect Euclidean symmetries and concentrate probability mass on thermodynamically favorable, mechanically stable structures. However, E(3)-equivariant diffusion models often reproduce biases from semi-empirical training data rather than capturing the equilibrium distribution of a high-fidelity Hamiltonian. While physics-based guidance can correct this, it faces two computational bottlenecks: expensive quantum-chemical evaluations (e.g., DFT) and the need to repeat such queries at every sampling step. We present Elign, a post-training framework that amortizes both costs. First, we replace expensive DFT evaluations with a faster, pretrained foundational machine-learning force field (MLFF) to provide physical signals. Second, we eliminate repeated run-time queries by shifting physical steering to the training phase. To achieve the second amortization, we formulate reverse diffusion as a reinforcement learning problem and introduce Force--Energy Disentangled Group Relative Policy Optimization (FED-GRPO) to fine-tune the denoising policy. FED-GRPO includes a potential-based energy reward and a force-based stability reward, which are optimized and group-normalized independently. Experiments show that Elign generates conformations with lower gold-standard DFT energies and forces, while improving stability. Crucially, inference remains as fast as unguided sampling, since no energy evaluations are required during generation. --- ### PowerGenie: Analytically-Guided Evolutionary Discovery of Superior Reconfigurable Power Converters Authors: Jian Gao, Yiwei Zou, Abhishek Pradhan, Wenhao Huang, Yumin Su, Kaiyuan Yang, Xuan Zhang Published: 2026-01-29 Categories: cs.LG, cs.AR arXiv: https://arxiv.org/abs/2601.21984v3 Discovering superior circuit topologies requires navigating an exponentially large design space-a challenge traditionally reserved for human experts. Existing AI methods either select from predefined templates or generate novel topologies at a limited scale without rigorous verification, leaving large-scale performance-driven discovery underexplored. We present PowerGenie, a framework for automated discovery of higher-performance reconfigurable power converters at scale. PowerGenie introduces: (1) an automated analytical framework that determines converter functionality and theoretical performance limits without component sizing or SPICE simulation, and (2) an evolutionary finetuning method that co-evolves a generative model with its training distribution through fitness selection and uniqueness verification. Unlike existing methods that suffer from mode collapse and overfitting, our approach achieves higher syntax validity, function validity, novelty rate, and figure-of-merit (FoM). PowerGenie discovers a novel 8-mode reconfigurable converter with 23% higher FoM than the best training topology. SPICE simulations confirm average absolute efficiency gains of 10% across 8 modes and up to 17% at a single mode. Code will be released upon publication. --- ### Investigating Batch Inference in a Sequential Monte Carlo Framework for Neural Networks Authors: Andrew Millard, Joshua Murphy, Peter Green, Simon Maskell Published: 2026-01-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.21983v1 Bayesian inference allows us to define a posterior distribution over the weights of a generic neural network (NN). Exact posteriors are usually intractable, in which case approximations can be employed. One such approximation - variational inference - is computationally efficient when using mini-batch stochastic gradient descent as subsets of the data are used for likelihood and gradient evaluations, though the approach relies on the selection of a variational distribution which sufficiently matches the form of the posterior. Particle-based methods such as Markov chain Monte Carlo and Sequential Monte Carlo (SMC) do not assume a parametric family for the posterior by typically require higher computational cost. These sampling methods typically use the full-batch of data for likelihood and gradient evaluations, which contributes to this computational expense. We explore several methods of gradually introducing more mini-batches of data (data annealing) into likelihood and gradient evaluations of an SMC sampler. We find that we can achieve up to $6\times$ faster training with minimal loss in accuracy on benchmark image classification problems using NNs. --- ### VERSA: Verified Event Data Format for Reliable Soccer Analytics Authors: Geonhee Jo, Mingu Kang, Kangmin Lee, Minho Lee, Pascal Bauer, Sang-Ki Ko Published: 2026-01-29 Categories: cs.AI, cs.DB arXiv: https://arxiv.org/abs/2601.21981v1 Event stream data is a critical resource for fine-grained analysis across various domains, including financial transactions, system operations, and sports. In sports, it is actively used for fine-grained analyses such as quantifying player contributions and identifying tactical patterns. However, the reliability of these models is fundamentally limited by inherent data quality issues that cause logical inconsistencies (e.g., incorrect event ordering or missing events). To this end, this study proposes VERSA (Verified Event Data Format for Reliable Soccer Analytics), a systematic verification framework that ensures the integrity of event stream data within the soccer domain. VERSA is based on a state-transition model that defines valid event sequences, thereby enabling the automatic detection and correction of anomalous patterns within the event stream data. Notably, our examination of event data from the K League 1 (2024 season), provided by Bepro, detected that 18.81% of all recorded events exhibited logical inconsistencies. Addressing such integrity issues, our experiments demonstrate that VERSA significantly enhances cross-provider consistency, ensuring stable and unified data representation across heterogeneous sources. Furthermore, we demonstrate that data refined by VERSA significantly improves the robustness and performance of a downstream task called VAEP, which evaluates player contributions. These results highlight that the verification process is highly effective in increasing the reliability of data-driven analysis. --- ### Evaluating the trustworthiness of the Fréchet Inception Distance with stochastic embedding representations Authors: Ciaran Bench, Vivek Desai, Carlijn Roozemond, Ruben van Engen, Spencer A. Thomas Published: 2026-01-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.21979v2 Feature embeddings acquired from pretrained models are widely used in medical applications of deep learning to assess the characteristics of datasets; e.g. to determine the quality of synthetic, generated medical images. The Fréchet Inception Distance (FID) is one popular synthetic image quality metric that relies on the assumption that the characteristic features of the data can be detected and encoded by an InceptionV3 model pretrained on ImageNet1K (natural images). While it is widely known that this makes it less effective for applications involving medical images, the extent to which the metric fails to capture meaningful differences in image characteristics is not obviously known. Here, we use Monte Carlo dropout to compute the predictive variance in the FID as well as a supplemental estimate of the predictive variance in the feature embedding model's latent representations. We show that the magnitudes of the predictive variances considered exhibit varying degrees of correlation with the extent to which test inputs (ImageNet1K validation set augmented at various strengths, and other external datasets) are out-of-distribution relative to its training data, providing some insight into the effectiveness of their use as indicators of the trustworthiness of the FID. --- ### Bridging Graph Structure and Knowledge-Guided Editing for Interpretable Temporal Knowledge Graph Reasoning Authors: Shiqi Fan, Quanming Yao, Hongyi Nie, Wentao Ma, Zhen Wang, Wen Hua Published: 2026-01-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.21978v1 Temporal knowledge graph reasoning (TKGR) aims to predict future events by inferring missing entities with dynamic knowledge structures. Existing LLM-based reasoning methods prioritize contextual over structural relations, struggling to extract relevant subgraphs from dynamic graphs. This limits structural information understanding, leading to unstructured, hallucination-prone inferences especially with temporal inconsistencies. To address this problem, we propose IGETR (Integration of Graph and Editing-enhanced Temporal Reasoning), a hybrid reasoning framework that combines the structured temporal modeling capabilities of Graph Neural Networks (GNNs) with the contextual understanding of LLMs. IGETR operates through a three-stage pipeline. The first stage aims to ground the reasoning process in the actual data by identifying structurally and temporally coherent candidate paths through a temporal GNN, ensuring that inference starts from reliable graph-based evidence. The second stage introduces LLM-guided path editing to address logical and semantic inconsistencies, leveraging external knowledge to refine and enhance the initial paths. The final stage focuses on integrating the refined reasoning paths to produce predictions that are both accurate and interpretable. Experiments on standard TKG benchmarks show that IGETR achieves state-of-the-art performance, outperforming strong baselines with relative improvements of up to 5.6% on Hits@1 and 8.1% on Hits@3 on the challenging ICEWS datasets. Additionally, we execute ablation studies and additional analyses confirm the effectiveness of each component. --- ### From Particles to Agents: Hallucination as a Metric for Cognitive Friction in Spatial Simulation Authors: Javier Argota Sánchez-Vaquerizo, Luis Borunda Monsivais Published: 2026-01-29 Categories: cs.HC, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2601.21977v1 Traditional architectural simulations (e.g. Computational Fluid Dynamics, evacuation, structural analysis) model elements as deterministic physics-based "particles" rather than cognitive "agents". To bridge this, we introduce \textbf{Agentic Environmental Simulations}, where Large Multimodal generative models actively predict the next state of spatial environments based on semantic expectation. Drawing on examples from accessibility-oriented AR pipelines and multimodal digital twins, we propose a shift from chronological time-steps to Episodic Spatial Reasoning, where simulations advance through meaningful, surprisal-triggered events. Within this framework we posit AI hallucinations as diagnostic tools. By formalizing the \textbf{Cognitive Friction} ($C_f$) it is possible to reveal "Phantom Affordances", i.e. semiotic ambiguities in built space. Finally, we challenge current HCI paradigms by treating environments as dynamic cognitive partners and propose a human-centered framework of cognitive orchestration for designing AI-driven simulations that preserve autonomy, affective clarity, and cognitive integrity. --- ### Mind the Gap: How Elicitation Protocols Shape the Stated-Revealed Preference Gap in Language Models Authors: Pranav Mahajan, Ihor Kendiukhov, Syed Hussain, Lydia Nottingham Published: 2026-01-29 Categories: cs.AI, cs.ET arXiv: https://arxiv.org/abs/2601.21975v2 Recent work identifies a stated-revealed (SvR) preference gap in language models (LMs): a mismatch between the values models endorse and the choices they make in context. Existing evaluations rely heavily on binary forced-choice prompting, which entangles genuine preferences with artifacts of the elicitation protocol. We systematically study how elicitation protocols affect SvR correlation across 24 LMs. Allowing neutrality and abstention during stated preference elicitation allows us to exclude weak signals, substantially improving Spearman's rank correlation ($ρ$) between volunteered stated preferences and forced-choice revealed preferences. However, further allowing abstention in revealed preferences drives $ρ$ to near-zero or negative values due to high neutrality rates. Finally, we find that system prompt steering using stated preferences during revealed preference elicitation does not reliably improve SvR correlation on AIRiskDilemmas. Together, our results show that SvR correlation is highly protocol-dependent and that preference elicitation requires methods that account for indeterminate preferences. --- ### Learning Decentralized LLM Collaboration with Multi-Agent Actor Critic Authors: Shuo Liu, Tianle Chen, Ryan Amiri, Christopher Amato Published: 2026-01-29 Categories: cs.AI, cs.DC, cs.MA arXiv: https://arxiv.org/abs/2601.21972v5 Recent work has explored optimizing LLM collaboration through Multi-Agent Reinforcement Learning (MARL). However, most MARL fine-tuning approaches rely on predefined execution protocols, which often require centralized execution. Decentralized LLM collaboration is more appealing in practice, as agents can run inference in parallel with flexible deployments. Also, current approaches use Monte Carlo methods for fine-tuning, which suffer from high variance and thus require more samples to train effectively. Actor-critic methods are prevalent in MARL for dealing with these issues; thus, we developed Multi-Agent Actor-Critic (MAAC) methods to optimize decentralized LLM collaboration. In this paper, we analyze when and why these MAAC methods are beneficial. We propose 2 MAAC approaches, \textbf{CoLLM-CC} with a \textbf{C}entralized \textbf{C}ritic and \textbf{CoLLM-DC} with \textbf{D}ecentralized \textbf{C}ritics. Our experiments across writing, coding, and game-playing domains show that Monte Carlo methods and CoLLM-DC can achieve performance comparable to CoLLM-CC in short-horizon and dense-reward settings. However, they both underperform CoLLM-CC on long-horizon or sparse-reward tasks, where Monte Carlo methods require substantially more samples and CoLLM-DC struggles to converge. --- ### Supervised Mixture-of-Experts for Surgical Grasping and Retraction Authors: Lorenzo Mazza, Ariel Rodriguez, Rayan Younis, Martin Lelis, Ortrun Hellig, Chenpan Li, Sebastian Bodenstedt, Martin Wagner, Stefanie Speidel Published: 2026-01-29 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2601.21971v2 Imitation learning has achieved remarkable success in robotic manipulation, yet its application to surgical robotics remains challenging due to data scarcity, constrained workspaces, and the need for an exceptional level of safety and predictability. We present a supervised Mixture-of-Experts (MoE) architecture designed for phase-structured surgical manipulation tasks, which can be added on top of any autonomous policy. Unlike prior surgical robot learning approaches that rely on multi-camera setups or thousands of demonstrations, we show that a lightweight action decoder policy like Action Chunking Transformer (ACT) can learn complex, long-horizon manipulation from less than 150 demonstrations using solely stereo endoscopic images, when equipped with our architecture. We evaluate our approach on the collaborative surgical task of bowel grasping and retraction, where a robot assistant interprets visual cues from a human surgeon, executes targeted grasping on deformable tissue, and performs sustained retraction. Our results show that generalist Vision Language Action models fail to acquire the task entirely, even under standard in-distribution conditions. Furthermore, while standard ACT achieves moderate success in-distribution, adopting a supervised MoE architecture significantly boosts its performance, yielding higher success rates in-distribution and demonstrating superior robustness in out-of-distribution scenarios, including novel grasp locations, reduced illumination, and partial occlusions. Notably, it generalizes to unseen testing viewpoints and also transfers zero-shot to ex vivo porcine tissue without additional training, offering a promising pathway toward in vivo deployment. To support this statement, we present qualitative preliminary results of policy roll-outs during in vivo porcine surgery. --- ### Token-Guard: Towards Token-Level Hallucination Control via Self-Checking Decoding Authors: Yifan Zhu, Huiqiang Rong, Haoran Luo Published: 2026-01-29 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2601.21969v2 Large Language Models (LLMs) often hallucinate, generating content inconsistent with the input. Retrieval-Augmented Generation (RAG) and Reinforcement Learning with Human Feedback (RLHF) can mitigate hallucinations but require resource-intensive retrieval or large-scale fine-tuning. Decoding-based methods are lighter yet lack explicit hallucination control. To address this, we present Token-Guard, a token-level hallucination control method based on self-checking decoding. Token-Guard performs internal verification at each reasoning step to detect hallucinated tokens before they propagate. Candidate fragments are further evaluated in a latent space with explicit hallucination risk scoring, while iterative pruning and regeneration dynamically correct detected errors. Experiments on HALU datasets show Token-Guard substantially reduces hallucinations and improves generation accuracy, offering a scalable, modular solution for reliable LLM outputs. Our code is publicly available. --- ### OVD: On-policy Verbal Distillation Authors: Jing Xiong, Hui Shen, Shansan Gong, Yuxin Cheng, Jianghan Shen, Chaofan Tao, Haochen Tan, Haoli Bai, Lifeng Shang, Ngai Wong Published: 2026-01-29 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.21968v2 Knowledge distillation transfers reasoning capabilities from large teachers to efficient students. However, token-level on-policy distillation (OPD) constrains student exploration and requires teacher token probabilities, precluding distillation from black-box teachers that provide only text outputs. We introduce On-policy Verbal Distillation (OVD), a framework that uses verbal scores from black-box teachers to rank student-generated sub-trajectories, retaining high-scoring ones and replacing low-scoring ones with teacher-generated continuations. We analyze when ranking induced by verbal scores can guide distribution approximation: under a density-ratio calibration condition on acceptance probabilities and bounded teacher-replacement error, we bound the approximation error between the resulting mixed trajectory distribution and a teacher-preferred target. On Web Q&A, OVD achieves 41.09% average EM with teacher feedback at inference, exceeding the strongest evaluated baseline by 5.89 percentage points. On AMC23, OVD-FR improves accuracy over RLVR by 10.0 percentage points (52.5% to 62.5%) after 600 training steps on 128 problems. Further experiments suggest that retaining student-generated prefixes helps preserve exploration and mitigate trajectory-level entropy collapse. OVD also improves training efficiency: resampling selected suffixes rather than entire responses reduces mean per-step training time by 10.2% in the 128-problem setting. Project page: https://menik1126.github.io/ovd-project-page/. --- ### The Energy Impact of Domain Model Design in Classical Planning Authors: Ilche Georgievski, Serhat Tekin, Marco Aiello Published: 2026-01-29 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2601.21967v1 AI research has traditionally prioritised algorithmic performance, such as optimising accuracy in machine learning or runtime in automated planning. The emerging paradigm of Green AI challenges this by recognising energy consumption as a critical performance dimension. Despite the high computational demands of automated planning, its energy efficiency has received little attention. This gap is particularly salient given the modular planning structure, in which domain models are specified independently of algorithms. On the other hand, this separation also enables systematic analysis of energy usage through domain model design. We empirically investigate how domain model characteristics affect the energy consumption of classical planners. We introduce a domain model configuration framework that enables controlled variation of features, such as element ordering, action arity, and dead-end states. Using five benchmark domains and five state-of-the-art planners, we analyse energy and runtime impacts across 32 domain variants per benchmark. Results demonstrate that domain-level modifications produce measurable energy differences across planners, with energy consumption not always correlating with runtime. --- ### From Tokens to Blocks: A Block-Diffusion Perspective on Molecular Generation Authors: Qianwei Yang, Dong Xu, Zhangfan Yang, Sisi Yuan, Zexuan Zhu, Jianqiang Li, Junkai Ji Published: 2026-01-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.21964v2 Drug discovery can be viewed as a combinatorial search over an immense chemical space, motivating the development of deep generative models for de novo molecular design. Among these, GPT-based molecular language models (MLM) have shown strong molecular design performance by learning chemical syntax and semantics from large-scale data. However, existing MLMs face two fundamental limitations: they inadequately capture the graph-structured nature of molecules when formulated as next-token prediction problems, and they typically lack explicit mechanisms for target-aware generation. Here, we propose SoftMol, a unified framework that co-designs molecular representation, model architecture, and search strategy for target-aware molecular generation. SoftMol introduces soft fragments, a rule-free block representation of SMILES that enables diffusion-native modeling, and develops SoftBD, the first block-diffusion molecular language model that combines local bidirectional diffusion with autoregressive generation under molecular structural constraints. To favor generated molecules with high drug-likeness and synthetic accessibility, SoftBD is trained on a carefully curated dataset named ZINC-Curated. SoftMol further integrates a gated Monte Carlo tree search to assemble fragments in a target-aware manner. Experimental results show that, compared with current state-of-the-art models, SoftMol achieves 100% chemical validity, improves binding affinity by 9.7%, yields a 2-3x increase in molecular diversity, and delivers a 6.6x speedup in inference efficiency. Code is available at https://github.com/szu-aicourse/softmol --- ### Industrialized Deception: The Collateral Effects of LLM-Generated Misinformation on Digital Ecosystems Authors: Alexander Loth, Martin Kappes, Marc-Oliver Pahl Published: 2026-01-29 Categories: cs.CY, cs.AI, cs.CL, cs.SI arXiv: https://arxiv.org/abs/2601.21963v2 Generative AI and misinformation research has evolved since our 2024 survey. This paper presents an updated perspective, transitioning from literature review to practical countermeasures. We report on changes in the threat landscape, including improved AI-generated content through Large Language Models (LLMs) and multimodal systems. Central to this work are our practical contributions: JudgeGPT, a platform for evaluating human perception of AI-generated news, and RogueGPT, a controlled stimulus generation engine for research. Together, these tools form an experimental pipeline for studying how humans perceive and detect AI-generated misinformation. Our findings show that detection capabilities have improved, but the competition between generation and detection continues. We discuss mitigation strategies including LLM-based detection, inoculation approaches, and the dual-use nature of generative AI. This work contributes to research addressing the adverse impacts of AI on information quality. --- ### How do Visual Attributes Influence Web Agents? A Comprehensive Evaluation of User Interface Design Factors Authors: Kuai Yu, Naicheng Yu, Han Wang, Rui Yang, Huan Zhang Published: 2026-01-29 Categories: cs.AI, cs.HC arXiv: https://arxiv.org/abs/2601.21961v2 Web agents have demonstrated strong performance on a wide range of web-based tasks. However, existing research on the effect of environmental variation has mostly focused on robustness to adversarial attacks, with less attention to agents' preferences in benign scenarios. Although early studies have examined how textual attributes influence agent behavior, a systematic understanding of how visual attributes shape agent decision-making remains limited. To address this, we introduce VAF, a controlled evaluation pipeline for quantifying how webpage Visual Attribute Factors influence web-agent decision-making. Specifically, VAF consists of three stages: (i) variant generation, which ensures the variants share identical semantics as the original item while only differ in visual attributes; (ii) browsing interaction, where agents navigate the page via scrolling and clicking the interested item, mirroring how human users browse online; (iii) validating through both click action and reasoning from agents, which we use the Target Click Rate and Target Mention Rate to jointly evaluate the effect of visual attributes. By quantitatively measuring the decision-making difference between the original and variant, we identify which visual attributes influence agents' behavior most. Extensive experiments, across 8 variant families (48 variants total), 5 real-world websites (including shopping, travel, and news browsing), and 4 representative web agents, show that background color contrast, item size, position, and card clarity have a strong influence on agents' actions, whereas font styling, text color, and item image clarity exhibit minor effects. --- ### Near-Optimal Private Tests for Simple and MLR Hypotheses Authors: Yu-Wei Chen, Raghu Pasupathy, Jordan Awan Published: 2026-01-29 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2601.21959v2 We develop a near-optimal testing procedure under the framework of Gaussian differential privacy for simple as well as one- and two-sided tests under monotone likelihood ratio conditions. Our mechanism is based on a private mean estimator with data-driven clamping bounds, whose population risk matches the private minimax rate up to logarithmic factors. Using this estimator, we construct private test statistics that achieve the same asymptotic relative efficiency as the non-private, most powerful tests while maintaining conservative type I error control. In addition to our theoretical results, our numerical experiments show that our private tests outperform competing DP methods and offer comparable power to the non-private most powerful tests, even at moderately small sample sizes and privacy loss budgets. --- ### PaddleOCR-VL-1.5: Towards a Multi-Task 0.9B VLM for Robust In-the-Wild Document Parsing Authors: Cheng Cui, Ting Sun, Suyin Liang, Tingquan Gao, Zelun Zhang, Jiaxuan Liu, Xueqing Wang, Changda Zhou, Hongen Liu, Manhui Lin, Yue Zhang, Yubo Zhang et al. Published: 2026-01-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.21957v2 We introduce PaddleOCR-VL-1.5, an upgraded model achieving a new state-of-the-art (SOTA) accuracy of 94.5% on OmniDocBench v1.5. To rigorously evaluate robustness against real-world physical distortions, including scanning, skew, warping, screen-photography, and illumination, we propose the Real5-OmniDocBench benchmark. Experimental results demonstrate that this enhanced model attains SOTA performance on the newly curated benchmark. Furthermore, we extend the model's capabilities by incorporating seal recognition and text spotting tasks, while remaining a 0.9B ultra-compact VLM with high efficiency. Code: https://github.com/PaddlePaddle/PaddleOCR --- ### Uncertainty-Aware Data-Based Method for Fast and Reliable Shape Optimization Authors: Yunjia Yang, Runze Li, Yufei Zhang, Haixin Chen Published: 2026-01-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.21956v1 Data-based optimization (DBO) offers a promising approach for efficiently optimizing shape for better aerodynamic performance by leveraging a pretrained surrogate model for offline evaluations during iterations. However, DBO heavily relies on the quality of the training database. Samples outside the training distribution encountered during optimization can lead to significant prediction errors, potentially misleading the optimization process. Therefore, incorporating uncertainty quantification into optimization is critical for detecting outliers and enhancing robustness. This study proposes an uncertainty-aware data-based optimization (UA-DBO) framework to monitor and minimize surrogate model uncertainty during DBO. A probabilistic encoder-decoder surrogate model is developed to predict uncertainties associated with its outputs, and these uncertainties are integrated into a model-confidence-aware objective function to penalize samples with large prediction errors during data-based optimization process. The UA-DBO framework is evaluated on two multipoint optimization problems aimed at improving airfoil drag divergence and buffet performance. Results demonstrate that UA-DBO consistently reduces prediction errors in optimized samples and achieves superior performance gains compared to original DBO. Moreover, compared to multipoint optimization based on full computational simulations, UA-DBO offers comparable optimization effectiveness while significantly accelerating optimization speed. --- ### From Generative Modeling to Clinical Classification: A GPT-Based Architecture for EHR Notes Authors: Fariba Afrin Irany, Sampson Akwafuo Published: 2026-01-29 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.21955v3 The increasing availability of unstructured clinical narratives in electronic health records (EHRs) has created new opportunities for automated disease characterization, cohort identification, and clinical decision support. However, modeling long, domain-specific clinical text remains challenging due to limited labeled data, severe class imbalance, and the high computational cost of adapting large pretrained language models. This study presents a GPT-based architecture for clinical text classification that adapts a pretrained decoder-only Transformer using a selective fine-tuning strategy. Rather than updating all model parameters, the majority of the GPT-2 backbone is frozen, and training is restricted to the final Transformer block, the final layer normalization, and a lightweight classification head. This approach substantially reduces the number of trainable parameters while preserving the representational capacity required to model complex clinical language. The proposed method is evaluated on radiology reports from the MIMIC-IV-Note dataset using uncertainty-aware CheXpert-style labels derived directly from report text. Experiments cover multiple problem formulations, including multi-label classification of radiographic findings, binary per-label classification under different uncertainty assumptions, and aggregate disease outcome prediction. Across varying dataset sizes, the model exhibits stable convergence behavior and strong classification performance, particularly in settings dominated by non-mention and negated findings. Overall, the results indicate that selective fine-tuning of pretrained generative language models provides an efficient and effective pathway for clinical text classification, enabling scalable adaptation to real-world EHR data while significantly reducing computational complexity. --- ### Diffusion Path Samplers via Sequential Monte Carlo Authors: James Matthew Young, Paula Cordero-Encinar, Sebastian Reich, Andrew Duncan, O. Deniz Akyildiz Published: 2026-01-29 Categories: stat.ML, cs.LG, stat.CO arXiv: https://arxiv.org/abs/2601.21951v2 We develop diffusion-based samplers for target distributions known up to a normalising constant. To this end, we rely on the well-known diffusion path that smoothly interpolates between a simple base distribution and the target, popularised by diffusion models. We tackle the score estimation problem by developing an efficient sequential Monte Carlo sampler that evolves auxiliary variables from conditional distributions along the path, providing principled score and density estimates for time-varying distributions. To control the variance of score estimates, we further propose practical control variate schedules that incur minimal overhead. We adapt this general framework to paths induced by the Ornstein-Uhlenbeck (OU) time-reversal process, stochastic interpolants, and diffusion annealed Langevin dynamics, outlining their trade-offs. Finally, we provide theoretical guarantees and empirically demonstrate the effectiveness of our method on several synthetic and real-world datasets. --- ### Embracing Aleatoric Uncertainty in Medical Multimodal Learning with Missing Modalities Authors: Linxiao Gong, Yang Liu, Lianlong Sun, Yulai Bi, Jing Liu, Xiaoguang Zhu Published: 2026-01-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.21950v1 Medical multimodal learning faces significant challenges with missing modalities prevalent in clinical practice. Existing approaches assume equal contribution of modality and random missing patterns, neglecting inherent uncertainty in medical data acquisition. In this regard, we propose the Aleatoric Uncertainty Modeling (AUM) that explicitly quantifies unimodal aleatoric uncertainty to address missing modalities. Specifically, AUM models each unimodal representation as a multivariate Gaussian distribution to capture aleatoric uncertainty and enable principled modality reliability quantification. To adaptively aggregate captured information, we develop a dynamic message-passing mechanism within a bipartite patient-modality graph using uncertainty-aware aggregation mechanism. Through this process, missing modalities are naturally accommodated, while more reliable information from available modalities is dynamically emphasized to guide representation generation. Our AUM framework achieves an improvement of 2.26% AUC-ROC on MIMIC-IV mortality prediction and 2.17% gain on eICU, outperforming existing state-of-the-art approaches. --- ### Deep Models, Shallow Alignment: Uncovering the Granularity Mismatch in Neural Decoding Authors: Yang Du, Siyuan Dai, Yonghao Song, Paul M. Thompson, Haoteng Tang, Liang Zhan Published: 2026-01-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.21948v2 Neural visual decoding is a central problem in brain-computer interface research, aiming to reconstruct human visual perception and to elucidate the structure of neural representations. Recent contrastive neural visual decoding methods commonly align neural signals with the final embeddings of pretrained vision encoders. However, such representations are optimized for high-level semantic invariance, whereas EEG/MEG signals contain information spanning multiple levels of visual abstraction, potentially creating a representational granularity mismatch. Motivated by prior evidence that brain representations correspond to multiple levels of the DNN hierarchy, we propose Shallow Alignment, a granularity-calibration framework that systematically explores intermediate visual representations as alignment targets for neural decoding. Extensive experiments across multiple benchmarks demonstrate that Shallow Alignment significantly outperforms standard final-layer alignment, with performance gains ranging from 22% to 58% across diverse vision backbones. Notably, our approach reveals a positive scaling trend in neural visual decoding, enabling decoding performance to improve consistently with the capacity of pre-trained vision backbones. We further conduct systematic empirical analyses to shed light on the mechanisms underlying the observed performance gains. Code is available at https://github.com/yangdu-neuroai/shallow-alignment. --- ### ToolWeaver: Weaving Collaborative Semantics for Scalable Tool Use in Large Language Models Authors: Bowen Fang, Wen Ye, Yunyue Su, Jinghao Zhang, Qiang Liu, Yesheng Liu, Xin Sun, Shu Wu, Jiabing Yang, Baole Wei, Liang Wang Published: 2026-01-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.21947v1 Prevalent retrieval-based tool-use pipelines struggle with a dual semantic challenge: their retrievers often employ encoders that fail to capture complex semantics, while the Large Language Model (LLM) itself lacks intrinsic tool knowledge from its natural language pretraining. Generative methods offer a powerful alternative by unifying selection and execution, tasking the LLM to directly learn and generate tool identifiers. However, the common practice of mapping each tool to a unique new token introduces substantial limitations: it creates a scalability and generalization crisis, as the vocabulary size explodes and each tool is assigned a semantically isolated token. This approach also creates a semantic bottleneck that hinders the learning of collaborative tool relationships, as the model must infer them from sparse co-occurrences of monolithic tool IDs within a vast library. To address these limitations, we propose ToolWeaver, a novel generative tool learning framework that encodes tools into hierarchical sequences. This approach makes vocabulary expansion logarithmic to the number of tools. Crucially, it enables the model to learn collaborative patterns from the dense co-occurrence of shared codes, rather than the sparse co-occurrence of monolithic tool IDs. We generate these structured codes through a novel tokenization process designed to weave together a tool's intrinsic semantics with its extrinsic co-usage patterns. These structured codes are then integrated into the LLM through a generative alignment stage, where the model is fine-tuned to produce the hierarchical code sequences. Evaluation results with nearly 47,000 tools show that ToolWeaver significantly outperforms state-of-the-art methods, establishing a more scalable, generalizable, and semantically-aware foundation for advanced tool-augmented agents. --- ### Dependence of Equilibrium Propagation Training Success on Network Architecture Authors: Qingshan Wang, Clara C. Wanjura, Florian Marquardt Published: 2026-01-29 Categories: cs.LG, cond-mat.dis-nn, cs.ET, cs.NE arXiv: https://arxiv.org/abs/2601.21945v1 The rapid rise of artificial intelligence has led to an unsustainable growth in energy consumption. This has motivated progress in neuromorphic computing and physics-based training of learning machines as alternatives to digital neural networks. Many theoretical studies focus on simple architectures like all-to-all or densely connected layered networks. However, these may be challenging to realize experimentally, e.g. due to connectivity constraints. In this work, we investigate the performance of the widespread physics-based training method of equilibrium propagation for more realistic architectural choices, specifically, locally connected lattices. We train an XY model and explore the influence of architecture on various benchmark tasks, tracking the evolution of spatially distributed responses and couplings during training. Our results show that sparse networks with only local connections can achieve performance comparable to dense networks. Our findings provide guidelines for further scaling up architectures based on equilibrium propagation in realistic settings. --- ### Clarity: The Flexibility-Interpretability Trade-Off in Sparsity-aware Concept Bottleneck Models Authors: Konstantinos P. Panousis, Diego Marcos Published: 2026-01-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.21944v3 The widespread adoption of deep learning models in computer vision has intensified concerns about interpretability. Despite strong performance, these models are often treated as black boxes, with limited systematic investigation of their decision-making processes. While many interpretability methods exist, objective evaluation of learned representations remains limited, particularly for approaches that rely on sparsity to "induce" interpretability. In this work, we investigate how modeling choices in Concept Bottleneck Models (CBMs) affect the semantic alignment of concept representations. We introduce Clarity, a post-hoc diagnostic measure that captures the interplay between downstream performance and the sparsity and precision of concept activations. Using an interpretability assessment framework grounded in datasets with ground-truth concept annotations, we evaluate both VLM- and attribute predictor-based CBMs across three amortized sparsity-inducing strategies ($\ell_1$, $\ell_0$, and Bernoulli-based), alongside several widely used sparsity-aware CBM methods from the literature. Our experiments reveal a critical flexibility-interpretability trade-off: a model's capacity to optimize task performance by deviating from semantic alignment. We demonstrate that under this trade-off, different methods exhibit markedly different behaviors even at comparable performance levels. Finally, we validate our framework through a principled human study, which confirms that Clarity aligns significantly more closely with human trust than standard evaluation metrics. --- ### Entropy-Based Dimension-Free Convergence and Loss-Adaptive Schedules for Diffusion Models Authors: Ahmad Aghapour, Erhan Bayraktar, Ziqing Zhang Published: 2026-01-29 Categories: cs.LG, cs.IT arXiv: https://arxiv.org/abs/2601.21943v1 Diffusion generative models synthesize samples by discretizing reverse-time dynamics driven by a learned score (or denoiser). Existing convergence analyses of diffusion models typically scale at least linearly with the ambient dimension, and sharper rates often depend on intrinsic-dimension assumptions or other geometric restrictions on the target distribution. We develop an alternative, information-theoretic approach to dimension-free convergence that avoids any geometric assumptions. Under mild assumptions on the target distribution, we bound KL divergence between the target and generated distributions by $O(H^2/K)$ (up to endpoint factors), where $H$ is the Shannon entropy and $K$ is the number of sampling steps. Moreover, using a reformulation of the KL divergence, we propose a Loss-Adaptive Schedule (LAS) for efficient discretization of reverse SDE which is lightweight and relies only on the training loss, requiring no post-training heavy computation. Empirically, LAS improves sampling quality over common heuristic schedules. --- ### Clustering in Deep Stochastic Transformers Authors: Lev Fedorov, Michaël E. Sander, Romuald Elie, Pierre Marion, Mathieu Laurière Published: 2026-01-29 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2601.21942v1 Transformers have revolutionized deep learning across various domains but understanding the precise token dynamics remains a theoretical challenge. Existing theories of deep Transformers with layer normalization typically predict that tokens cluster to a single point; however, these results rely on deterministic weight assumptions, which fail to capture the standard initialization scheme in Transformers. In this work, we show that accounting for the intrinsic stochasticity of random initialization alters this picture. More precisely, we analyze deep Transformers where noise arises from the random initialization of value matrices. Under diffusion scaling and token-wise RMS normalization, we prove that, as the number of Transformer layers goes to infinity, the discrete token dynamics converge to an interacting-particle system on the sphere where tokens are driven by a \emph{common} matrix-valued Brownian noise. In this limit, we show that initialization noise prevents the collapse to a single cluster predicted by deterministic models. For two tokens, we prove a phase transition governed by the interaction strength and the token dimension: unlike deterministic attention flows, antipodal configurations become attracting with positive probability. Numerical experiments confirm the predicted transition, reveal that antipodal formations persist for more than two tokens, and demonstrate that suppressing the intrinsic noise degrades accuracy. --- ### Robust Multimodal Representation Learning in Healthcare Authors: Xiaoguang Zhu, Linxiao Gong, Lianlong Sun, Yang Liu, Haoyu Wang, Jing Liu Published: 2026-01-29 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2601.21941v1 Medical multimodal representation learning aims to integrate heterogeneous data into unified patient representations to support clinical outcome prediction. However, real-world medical datasets commonly contain systematic biases from multiple sources, which poses significant challenges for medical multimodal representation learning. Existing approaches typically focus on effective multimodal fusion, neglecting inherent biased features that affect the generalization ability. To address these challenges, we propose a Dual-Stream Feature Decorrelation Framework that identifies and handles the biases through structural causal analysis introduced by latent confounders. Our method employs a causal-biased decorrelation framework with dual-stream neural networks to disentangle causal features from spurious correlations, utilizing generalized cross-entropy loss and mutual information minimization for effective decorrelation. The framework is model-agnostic and can be integrated into existing medical multimodal learning methods. Comprehensive experiments on MIMIC-IV, eICU, and ADNI datasets demonstrate consistent performance improvements. --- ### BookNet: Book Image Rectification via Cross-Page Attention Network Authors: Shaokai Liu, Hao Feng, Bozhi Luan, Min Hou, Jiajun Deng, Wengang Zhou Published: 2026-01-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.21938v1 Book image rectification presents unique challenges in document image processing due to complex geometric distortions from binding constraints, where left and right pages exhibit distinctly asymmetric curvature patterns. However, existing single-page document image rectification methods fail to capture the coupled geometric relationships between adjacent pages in books. In this work, we introduce BookNet, the first end-to-end deep learning framework specifically designed for dual-page book image rectification. BookNet adopts a dual-branch architecture with cross-page attention mechanisms, enabling it to estimate warping flows for both individual pages and the complete book spread, explicitly modeling how left and right pages influence each other. Moreover, to address the absence of specialized datasets, we present Book3D, a large-scale synthetic dataset for training, and Book100, a comprehensive real-world benchmark for evaluation. Extensive experiments demonstrate that BookNet outperforms existing state-of-the-art methods on book image rectification. Code and dataset will be made publicly available. --- ### Retrieval-Infused Reasoning Sandbox: A Benchmark for Decoupling Retrieval and Reasoning Capabilities Authors: Shuangshuang Ying, Zheyu Wang, Yunjian Peng, Jin Chen, Yuhao Wu, Hongbin Lin, Dingyu He, Siyi Liu, Gengchen Yu, YinZhu Piao, Yuchen Wu, Xin Gui et al. Published: 2026-01-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.21937v2 Despite strong performance on existing benchmarks, it remains unclear whether large language models can reason over genuinely novel scientific information. Most evaluations score end-to-end RAG pipelines, where reasoning is confounded with retrieval and toolchain choices, and the signal is further contaminated by parametric memorization and open-web volatility. We introduce DeR2, a controlled deep-research sandbox that isolates document-grounded reasoning while preserving core difficulties of deep search: multi-step synthesis, denoising, and evidence-based conclusion making. DeR2 decouples evidence access from reasoning via four regimes--Instruction-only, Concepts (gold concepts without documents), Related-only (only relevant documents), and Full-set (relevant documents plus topically related distractors)--yielding interpretable regime gaps that operationalize retrieval loss vs. reasoning loss and enable fine-grained error attribution. To prevent parametric leakage, we apply a two-phase validation that requires parametric failure without evidence while ensuring oracle-concept solvability. To ensure reproducibility, each instance provides a frozen document library (drawn from 2023-2025 theoretical papers) with expert-annotated concepts and validated rationales. Experiments across a diverse set of state-of-the-art foundation models reveal substantial variation and significant headroom: some models exhibit mode-switch fragility, performing worse with the Full-set than with Instruction-only, while others show structural concept misuse, correctly naming concepts but failing to execute them as procedures. --- ### AgenticSimLaw: A Juvenile Courtroom Multi-Agent Debate Simulation for Explainable High-Stakes Tabular Decision Making Authors: Jon Chun, Kathrine Elkins, Yong Suk Lee Published: 2026-01-29 Categories: cs.AI arXiv: https://arxiv.org/abs/2601.21936v1 We introduce AgenticSimLaw, a role-structured, multi-agent debate framework that provides transparent and controllable test-time reasoning for high-stakes tabular decision-making tasks. Unlike black-box approaches, our courtroom-style orchestration explicitly defines agent roles (prosecutor, defense, judge), interaction protocols (7-turn structured debate), and private reasoning strategies, creating a fully auditable decision-making process. We benchmark this framework on young adult recidivism prediction using the NLSY97 dataset, comparing it against traditional chain-of-thought (CoT) prompting across almost 90 unique combinations of models and strategies. Our results demonstrate that structured multi-agent debate provides more stable and generalizable performance compared to single-agent reasoning, with stronger correlation between accuracy and F1-score metrics. Beyond performance improvements, AgenticSimLaw offers fine-grained control over reasoning steps, generates complete interaction transcripts for explainability, and enables systematic profiling of agent behaviors. While we instantiate this framework in the criminal justice domain to stress-test reasoning under ethical complexity, the approach generalizes to any deliberative, high-stakes decision task requiring transparency and human oversight. This work addresses key LLM-based multi-agent system challenges: organization through structured roles, observability through logged interactions, and responsibility through explicit non-deployment constraints for sensitive domains. Data, results, and code will be available on github.com under the MIT license. --- ### Just Noticeable Difference Modeling for Deep Visual Features Authors: Rui Zhao, Wenrui Li, Lin Zhu, Yajing Zheng, Weisi Lin Published: 2026-01-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.21933v1 Deep visual features are increasingly used as the interface in vision systems, motivating the need to describe feature characteristics and control feature quality for machine perception. Just noticeable difference (JND) characterizes the maximum imperceptible distortion for images under human or machine vision. Extending it to deep visual features naturally meets the above demand by providing a task-aligned tolerance boundary in feature space, offering a practical reference for controlling feature quality under constrained resources. We propose FeatJND, a task-aligned JND formulation that predicts the maximum tolerable per-feature perturbation map while preserving downstream task performance. We propose a FeatJND estimator at standardized split points and validate it across image classification, detection, and instance segmentation. Under matched distortion strength, FeatJND-based distortions consistently preserve higher task performance than unstructured Gaussian perturbations, and attribution visualizations suggest FeatJND can suppress non-critical feature regions. As an application, we further apply FeatJND to token-wise dynamic quantization and show that FeatJND-guided step-size allocation yields clear gains over random step-size permutation and global uniform step size under the same noise budget. Our code will be released after publication. --- ### LoRIF: Low-Rank Influence Functions for Scalable Training Data Attribution Authors: Shuangqi Li, Hieu Le, Jingyi Xu, Mathieu Salzmann Published: 2026-01-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2601.21929v2 Training data attribution (TDA) identifies which training examples most influenced a model's prediction. Influence function methods are a theoretically grounded family of TDA methods and exploit gradients. To overcome the scalability challenge arising from gradient computation, the most popular strategy is random projection (e.g., TRAK, LoGRA). However, this still faces two bottlenecks when scaling to large training sets and high-quality attribution: \emph{(i)} storing and loading projected per-example gradients for all $N$ training examples, where query latency is dominated by I/O; and \emph{(ii)} forming the $D \times D$ inverse Hessian approximation, which costs $O(D^2)$ memory. Both bottlenecks scale with the projection dimension $D$, yet increasing $D$ is necessary for attribution quality -- creating a quality--scalability tradeoff. We introduce \textbf{LoRIF} (\textbf{Lo}w-\textbf{R}ank \textbf{I}nfluence \textbf{F}unctions), which exploits low-rank structures of gradient to address both bottlenecks. First, we store rank-$c$ factors of projected per-example gradients rather than full matrices, reducing storage and query-time I/O from $O(D)$ to $O(c\sqrt{D})$ per layer per sample. Second, we use truncated SVD with the Woodbury identity to approximate the inverse Hessian term in an $r$-dimensional subspace, reducing memory from $O(D^2)$ to $O(Dr)$. On models from 0.1B to 70B parameters trained on datasets with millions of examples, LoRIF achieves up to 20$\times$ storage reduction and query-time speedup compared to LoGRA, while matching or exceeding its attribution quality. LoRIF makes gradient-based TDA practical at frontier scale. --- ### SONIC: Segmented Optimized Nexus for Information Compression in Key-Value Caching Authors: Hong Chen, Xiang Liu, Bo Wang, Yuxuan Fan, Yuanlin Chu, Zongluo Li, Xiaowen Chu, Xuming Hu Published: 2026-01-29 Categories: cs.CL arXiv: https://arxiv.org/abs/2601.21927v1 The linear growth of Key-Value (KV) cache remains a bottleneck for multi-turn LLM deployment. Existing KV cache compression methods often fail to account for the structural properties of multi-turn dialogues, relying on heuristic eviction that risks losing critical context. We propose \textbf{SONIC}, a learning-based framework that compresses historical segments into compact and semantically rich \textbf{Nexus} tokens. By integrating dynamic budget training, SONIC allows flexible adaptation to varying memory constraints without retraining. Experiments show that at compression ratios of 80\% and 50\%, SONIC consistently outperforms baselines such as H2O and StreamingLLM on four diverse multi-turn benchmarks. Specifically, on the widely used MTBench101 benchmark, SONIC achieves an average score improvement of 35.55\% over state-of-the-art baselines, validating its effectiveness in sustaining coherent multi-turn dialogues. Furthermore, SONIC enhances deployment efficiency, accelerating the overall inference process by 50.1\% compared to full-context generation. --- ### One-Step Bellman Alignment Enables Provably Efficient Transfer in Online RL Authors: Elynn Chen, Enpei Zhang, Jinhang Chai, Yujun Yan Published: 2026-01-29 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2601.21924v2 We study online transfer reinforcement learning (RL) in episodic Markov decision processes, where experience from related source tasks is available during learning on a target task. A fundamental difficulty is that task similarity is typically defined in terms of rewards or transitions, whereas online RL algorithms operate on Bellman regression targets. As a result, naively reusing source Bellman updates introduces systematic bias and invalidates regret guarantees. We identify one-step Bellman alignment as the correct abstraction for transfer in online RL and propose re-weighted targeting (RWT), an operator-level correction that retargets continuation values and compensates for transition mismatch via a change of measure. RWT reduces task mismatch to a fixed one-step correction and enables statistically sound reuse of source data. This alignment yields a two-stage RWT $Q$-learning framework that separates variance reduction from bias correction. Under RKHS function approximation, we establish regret bounds that scale with the complexity of the task shift rather than the target MDP. We further show the required density ratios admit a constructive RKHS estimator with finite-sample guarantees, and empirically validate robustness to estimated and mis-specified ratios. Empirical results in both tabular and neural network settings demonstrate consistent improvements over single-task learning and naïve pooling, highlighting Bellman alignment as a model-agnostic transfer principle for online RL. --- ### Zero-Shot Video Restoration and Enhancement with Assistance of Video Diffusion Models Authors: Cong Cao, Huanjing Yue, Shangbin Xie, Xin Liu, Jingyu Yang Published: 2026-01-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.21922v1 Although diffusion-based zero-shot image restoration and enhancement methods have achieved great success, applying them to video restoration or enhancement will lead to severe temporal flickering. In this paper, we propose the first framework that utilizes the rapidly-developed video diffusion model to assist the image-based method in maintaining more temporal consistency for zero-shot video restoration and enhancement. We propose homologous latents fusion, heterogenous latents fusion, and a COT-based fusion ratio strategy to utilize both homologous and heterogenous text-to-video diffusion models to complement the image method. Moreover, we propose temporal-strengthening post-processing to utilize the image-to-video diffusion model to further improve temporal consistency. Our method is training-free and can be applied to any diffusion-based image restoration and enhancement methods. Experimental results demonstrate the superiority of the proposed method. --- ### From Future of Work to Future of Workers: Addressing Asymptomatic AI Harms for Dignified Human-AI Interaction Authors: Upol Ehsan, Samir Passi, Koustuv Saha, Todd McNutt, Mark O. Riedl, Sara Alcorn Published: 2026-01-29 Categories: cs.HC, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2601.21920v1 In the future of work discourse, AI is touted as the ultimate productivity amplifier. Yet, beneath the efficiency gains lie subtle erosions of human expertise and agency. This paper shifts focus from the future of work to the future of workers by navigating the AI-as-Amplifier Paradox: AI's dual role as enhancer and eroder, simultaneously strengthening performance while eroding underlying expertise. We present a year-long study on the longitudinal use of AI in a high-stakes workplace among cancer specialists. Initial operational gains hid ``intuition rust'': the gradual dulling of expert judgment. These asymptomatic effects evolved into chronic harms, such as skill atrophy and identity commoditization. Building on these findings, we offer a framework for dignified Human-AI interaction co-constructed with professional knowledge workers facing AI-induced skill erosion without traditional labor protections. The framework operationalizes sociotechnical immunity through dual-purpose mechanisms that serve institutional quality goals while building worker power to detect, contain, and recover from skill erosion, and preserve human identity. Evaluated across healthcare and software engineering, our work takes a foundational step toward dignified human-AI interaction futures by balancing productivity with the preservation of human expertise. --- ### Self-Compression of Chain-of-Thought via Multi-Agent Reinforcement Learning Authors: Yiqun Chen, Jinyuan Feng, Wei Yang, Meizhi Zhong, Zhengliang Shi, Rui Li, Xiaochi Wei, Yan Gao, Yi Wu, Yao Hu, Zhiqiang Pu, Jiaxin Mao Published: 2026-01-29 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2601.21919v1 The inference overhead induced by redundant reasoning undermines the interactive experience and severely bottlenecks the deployment of Large Reasoning Models. Existing reinforcement learning (RL)-based solutions tackle this problem by coupling a length penalty with outcome-based rewards. This simplistic reward weighting struggles to reconcile brevity with accuracy, as enforcing brevity may compromise critical reasoning logic. In this work, we address this limitation by proposing a multi-agent RL framework that selectively penalizes redundant chunks, while preserving essential reasoning logic. Our framework, Self-Compression via MARL (SCMA), instantiates redundancy detection and evaluation through two specialized agents: \textbf{a Segmentation Agent} for decomposing the reasoning process into logical chunks, and \textbf{a Scoring Agent} for quantifying the significance of each chunk. The Segmentation and Scoring agents collaboratively define an importance-weighted length penalty during training, incentivizing \textbf{a Reasoning Agent} to prioritize essential logic without introducing inference overhead during deployment. Empirical evaluations across model scales demonstrate that SCMA reduces response length by 11.1\% to 39.0\% while boosting accuracy by 4.33\% to 10.02\%. Furthermore, ablation studies and qualitative analysis validate that the synergistic optimization within the MARL framework fosters emergent behaviors, yielding more powerful LRMs compared to vanilla RL paradigms. --- ### On Approximate Computation of Critical Points Authors: Amir Ali Ahmadi, Georgina Hall Published: 2026-01-29 Categories: math.OC, cs.CC, cs.LG, math.NA, stat.ML arXiv: https://arxiv.org/abs/2601.21917v1 We show that computing even very coarse approximations of critical points is intractable for simple classes of nonconvex functions. More concretely, we prove that if there exists a polynomial-time algorithm that takes as input a polynomial in $n$ variables of constant degree (as low as three) and outputs a point whose gradient has Euclidean norm at most $2^n$ whenever the polynomial has a critical point, then P=NP. The algorithm is permitted to return an arbitrary point when no critical point exists. We also prove hardness results for approximate computation of critical points under additional structural assumptions, including settings in which existence and uniqueness of a critical point are guaranteed, the function is lower bounded, and approximation is measured in terms of distance to a critical point. Overall, our results stand in contrast to the commonly-held belief that, in nonconvex optimization, approximate computation of critical points is a tractable task. --- ### JADE: Bridging the Strategic-Operational Gap in Dynamic Agentic RAG Authors: Yiqun Chen, Erhan Zhang, Tianyi Hu, Shijie Wang, Zixuan Yang, Meizhi Zhong, Xiaochi Wei, Yan Gao, Yi Wu, Yao Hu, Jiaxin Mao Published: 2026-01-29 Categories: cs.AI, cs.CL, cs.IR arXiv: https://arxiv.org/abs/2601.21916v1 The evolution of Retrieval-Augmented Generation (RAG) has shifted from static retrieval pipelines to dynamic, agentic workflows where a central planner orchestrates multi-turn reasoning. However, existing paradigms face a critical dichotomy: they either optimize modules jointly within rigid, fixed-graph architectures, or empower dynamic planning while treating executors as frozen, black-box tools. We identify that this \textit{decoupled optimization} creates a ``strategic-operational mismatch,'' where sophisticated planning strategies fail to materialize due to unadapted local executors, often leading to negative performance gains despite increased system complexity. In this paper, we propose \textbf{JADE} (\textbf{J}oint \textbf{A}gentic \textbf{D}ynamic \textbf{E}xecution), a unified framework for the joint optimization of planning and execution within dynamic, multi-turn workflows. By modeling the system as a cooperative multi-agent team unified under a single shared backbone, JADE enables end-to-end learning driven by outcome-based rewards. This approach facilitates \textit{co-adaptation}: the planner learns to operate within the capability boundaries of the executors, while the executors evolve to align with high-level strategic intent. Empirical results demonstrate that JADE transforms disjoint modules into a synergistic system, yielding remarkable performance improvements via joint optimization and enabling a flexible balance between efficiency and effectiveness through dynamic workflow orchestration. --- ### VideoAesBench: Benchmarking the Video Aesthetics Perception Capabilities of Large Multimodal Models Authors: Yunhao Li, Sijing Wu, Zhilin Gao, Zicheng Zhang, Qi Jia, Huiyu Duan, Xiongkuo Min, Guangtao Zhai Published: 2026-01-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.21915v2 Large multimodal models (LMMs) have demonstrated outstanding capabilities in various visual perception tasks, which has in turn made the evaluation of LMMs significant. However, the capability of video aesthetic quality assessment, which is a fundamental ability for human, remains underexplored for LMMs. To address this, we introduce VideoAesBench, a comprehensive benchmark for evaluating LMMs' understanding of video aesthetic quality. VideoAesBench has several significant characteristics: (1) Diverse content including 1,804 videos from multiple video sources including user-generated (UGC), AI-generated (AIGC), compressed, robotic-generated (RGC), and game videos. (2) Multiple question formats containing traditional single-choice questions, multi-choice questions, True or False questions, and a novel open-ended questions for video aesthetics description. (3) Holistic video aesthetics dimensions including visual form related questions from 5 aspects, visual style related questions from 4 aspects, and visual affectiveness questions from 3 aspects. Based on VideoAesBench, we benchmark 23 open-source and commercial large multimodal models. Our findings show that current LMMs only contain basic video aesthetics perception ability, their performance remains incomplete and imprecise. We hope our VideoAesBench can be served as a strong testbed and offer insights for explainable video aesthetics assessment. The data will be released on https://github.com/michaelliyunhao/VideoAesBench --- ### ProRAG: Process-Supervised Reinforcement Learning for Retrieval-Augmented Generation Authors: Zhao Wang, Ziliang Zhao, Zhicheng Dou Published: 2026-01-29 Categories: cs.AI, cs.CL, cs.IR arXiv: https://arxiv.org/abs/2601.21912v1 Reinforcement learning (RL) has become a promising paradigm for optimizing Retrieval-Augmented Generation (RAG) in complex reasoning tasks. However, traditional outcome-based RL approaches often suffer from reward sparsity and inefficient credit assignment, as coarse-grained scalar rewards fail to identify specific erroneous steps within long-horizon trajectories. This ambiguity frequently leads to "process hallucinations", where models reach correct answers through flawed logic or redundant retrieval steps. Although recent process-aware approaches attempt to mitigate this via static preference learning or heuristic reward shaping, they often lack the on-policy exploration capabilities required to decouple step-level credit from global outcomes. To address these challenges, we propose ProRAG, a process-supervised reinforcement learning framework designed to integrate learned step-level supervision into the online optimization loop. Our framework consists of four stages: (1) Supervised Policy Warmup to initialize the model with a structured reasoning format; (2) construction of an MCTS-based Process Reward Model (PRM) to quantify intermediate reasoning quality; (3) PRM-Guided Reasoning Refinement to align the policy with fine-grained process preferences; and (4) Process-Supervised Reinforcement Learning with a dual-granularity advantage mechanism. By aggregating step-level process rewards with global outcome signals, ProRAG provides precise feedback for every action. Extensive experiments on five multi-hop reasoning benchmarks demonstrate that ProRAG achieves superior overall performance compared to strong outcome-based and process-aware RL baselines, particularly on complex long-horizon tasks, validating the effectiveness of fine-grained process supervision. The code and model are available at https://github.com/lilinwz/ProRAG. --- ### From Meta-Thought to Execution: Cognitively Aligned Post-Training for Generalizable and Reliable LLM Reasoning Authors: Shaojie Wang, Liang Zhang Published: 2026-01-29 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2601.21909v2 Current LLM post-training methods optimize complete reasoning trajectories through Supervised Fine-Tuning (SFT) followed by outcome-based Reinforcement Learning (RL). While effective, a closer examination reveals a fundamental gap: this approach does not align with how humans actually solve problems. Human cognition naturally decomposes problem-solving into two distinct stages: first acquiring abstract strategies (i.e., meta-knowledge) that generalize across problems, then adapting them to specific instances. In contrast, by treating complete trajectories as basic units, current methods are inherently problem-centric, entangling abstract strategies with problem-specific execution. To address this misalignment, we propose a cognitively-inspired framework that explicitly mirrors the two-stage human cognitive process. Specifically, Chain-of-Meta-Thought CoMT focuses supervised learning on abstract reasoning patterns without specific executions, enabling acquisition of generalizable strategies. Confidence-Calibrated Reinforcement Learning (CCRL) then optimizes task adaptation via confidence-aware rewards on intermediate steps, preventing overconfident errors from cascading and improving execution reliability. Experiments across four models and ten benchmarks show 2.10% and 3.86% improvements in-distribution and out-of-distribution respectively over standard methods, while remaining highly robust to variations in teacher model selection, optimization methods, and symbolic perturbations. --- ### Beyond Global Alignment: Fine-Grained Motion-Language Retrieval via Pyramidal Shapley-Taylor Learning Authors: Hanmo Chen, Guangtao Lyu, Chenghao Xu, Jiexi Yan, Xu Yang, Cheng Deng Published: 2026-01-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2601.21904v3 As a foundational task in human-centric cross-modal intelligence, motion-language retrieval aims to bridge the semantic gap between natural language and human motion, enabling intuitive motion analysis, yet existing approaches predominantly focus on aligning entire motion sequences with global textual representations. This global-centric paradigm overlooks fine-grained interactions between local motion segments and individual body joints and text tokens, inevitably leading to suboptimal retrieval performance. To address this limitation, we draw inspiration from the pyramidal process of human motion perception (from joint dynamics to segment coherence, and finally to holistic comprehension) and propose a novel Pyramidal Shapley-Taylor (PST) learning framework for fine-grained motion-language retrieval. Specifically, the framework decomposes human motion into temporal segments and spatial body joints, and learns cross-modal correspondences through progressive joint-wise and segment-wise alignment in a pyramidal fashion, effectively capturing both local semantic details and hierarchical structural relationships. Extensive experiments on multiple public benchmark datasets demonstrate that our approach significantly outperforms state-of-the-art methods, achieving precise alignment between motion segments and body joints and their corresponding text tokens. The code of this work will be released upon acceptance. ---