# Research Papers — 2026-01 · bundle page 8 of 189 > Full-text markdown of papers #351–400 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/8 [← Previous bundle](https://aipod.dev/papers/archive/2026-01/page/7.md) [Next bundle →](https://aipod.dev/papers/archive/2026-01/page/9.md) ### PA-MIL: Phenotype-Aware Multiple Instance Learning Guided by Language Prompting and Genotype-to-Phenotype Relationships Authors: Zekang Yang, Hong Liu, Xiangdong Wang Published: 2026-01-30 Categories: cs.LG, cs.AI, q-bio.QM arXiv: https://arxiv.org/abs/2602.02558v1 Deep learning has been extensively researched in the analysis of pathology whole-slide images (WSIs). However, most existing methods are limited to providing prediction interpretability by locating the model's salient areas in a post-hoc manner, failing to offer more reliable and accountable explanations. In this work, we propose Phenotype-Aware Multiple Instance Learning (PA-MIL), a novel ante-hoc interpretable framework that identifies cancer-related phenotypes from WSIs and utilizes them for cancer subtyping. To facilitate PA-MIL in learning phenotype-aware features, we 1) construct a phenotype knowledge base containing cancer-related phenotypes and their associated genotypes. 2) utilize the morphological descriptions of phenotypes as language prompting to aggregate phenotype-related features. 3) devise the Genotype-to-Phenotype Neural Network (GP-NN) grounded in genotype-to-phenotype relationships, which provides multi-level guidance for PA-MIL. Experimental results on multiple datasets demonstrate that PA-MIL achieves competitive performance compared to existing MIL methods while offering improved interpretability. PA-MIL leverages phenotype saliency as evidence and, using a linear classifier, achieves competitive results compared to state-of-the-art methods. Additionally, we thoroughly analyze the genotype-phenotype relationships, as well as cohort-level and case-level interpretability, demonstrating the reliability and accountability of PA-MIL. --- ### The Alignment Curse: Modality Alignment Supercharges Audio Attacks via Text Transfer Authors: Yupeng Chen, Junchi Yu, Aoxi Liu, Baoyuan Wu, Philip Torr, Adel Bibi Published: 2026-01-30 Categories: cs.LG, cs.AI, cs.SD arXiv: https://arxiv.org/abs/2602.02557v2 Recent advances in end-to-end trained omni-models have substantially improved audio capabilities by strengthening text-audio modality alignment. However, whether such alignment inadvertently facilitates the transfer of safety vulnerabilities across modalities remains underexplored. This question is critical as text-based jailbreak attacks are considerably more mature than audio-based ones; if they transfer systematically, current audio safety evaluations may underestimate risks originating from the text modality. In this paper, we introduce the Alignment Curse, a formally characterized and empirically validated principle showing that stronger modality alignment enables more effective transfer of attacks from text to audio, revealing a fundamental tension between capability and safety. Motivated by this principle, we conduct a comprehensive black-box evaluation of three attack categories on recent omni-models (e.g., Qwen2.5-Omni, Qwen3-Omni): text attacks, text-transferred audio attacks, and audio attacks. We find that text-transferred audio attacks perform comparably to, and often better than, audio-based attacks, exhibiting a clear advantage under audio-only access. This suggests that text-based vulnerabilities play a pivotal role in shaping audio safety risks. Finally, we empirically analyze the relationship between modality alignment and transfer effectiveness across attack methods and models, observing consistent support for the Alignment Curse: tighter modality alignment leads to more effective cross-modality attack transfer. --- ### Beyond Experience Retrieval: Learning to Generate Utility-Optimized Structured Experience for Frozen LLMs Authors: Xuancheng Li, Haitao Li, Yujia Zhou, Yiqun Liu, Qingyao Ai Published: 2026-01-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.02556v2 Large language models (LLMs) are largely static and often redo reasoning or repeat mistakes. Prior experience reuse typically relies on external retrieval, which is similarity-based, can introduce noise, and adds latency. We introduce SEAM (Structured Experience Adapter Module), a lightweight, executor-specific plug-in that stores experience in its parameters and generates a structured, instance-tailored experience entry in a single forward pass to guide a frozen LLM executor. SEAM is trained for utility via executor rollouts and GRPO while keeping the executor frozen, and it can be further improved after deployment with supervised fine-tuning on logged successful trajectories. Experiments on mathematical reasoning benchmarks show consistent accuracy gains across executors with low overhead. Extensive ablations and analyses further elucidate the mechanisms underlying SEAM's effectiveness and robustness. --- ### Learning to Explore with Parameter-Space Noise: A Deep Dive into Parameter-Space Noise for Reinforcement Learning with Verifiable Rewards Authors: Bizhe Bai, Xinyue Wang, Peng Ye, Tao Chen Published: 2026-01-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.02555v3 Reinforcement Learning with Verifiable Rewards (RLVR) improves LLM reasoning, yet growing evidence indicates an exploration ceiling: it often reweights existing solution traces rather than discovering new strategies, limiting gains under large sampling budgets (e.g., pass-at-256). We address this limitation with PSN-RLVR, which perturbs policy parameters before rollout generation to induce temporally consistent, trajectory-level exploration that better preserves long-horizon chain-of-thought coherence than action-space noise. To mitigate the resulting sampling-update mismatch, we incorporate truncated importance sampling (TIS). To avoid expensive KL-based adaptive noise control, we propose a computationally efficient real-time adaptive noise scheduler driven by a lightweight surrogate that combines semantic diversity with normalized self-certainty. Instantiated on GRPO, a widely used RLVR method, PSN-GRPO consistently expands the effective reasoning capability boundary across multiple mathematical reasoning benchmarks and model families, yielding higher pass-at-k under large sampling budgets and outperforming prior exploration-oriented RLVR methods (e.g., Pass-at-k-style training) while remaining orthogonal and thus composable for additional gains. --- ### BatCoder: Self-Supervised Bidirectional Code-Documentation Learning via Back-Translation Authors: Jingwen Xu, Yiyang Lu, Zisu Huang, Changze Lv, Xiaohua Wang, Shizheng Li, Zhibo Xu, Zhengkang Guo, Zhengyuan Wang, Muzhao Tian, Xuanjing Huang, Xiaoqing Zheng Published: 2026-01-30 Categories: cs.LG, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2602.02554v1 Training LLMs for code-related tasks typically depends on high-quality code-documentation pairs, which are costly to curate and often scarce for niche programming languages. We introduce BatCoder, a self-supervised reinforcement learning framework designed to jointly optimize code generation and documentation production. BatCoder employs a back-translation strategy: a documentation is first generated from code, and then the generated documentation is used to reconstruct the original code. The semantic similarity between the original and reconstructed code serves as an implicit reward, enabling reinforcement learning to improve the model's performance both in generating code from documentation and vice versa. This approach allows models to be trained using only code, substantially increasing the available training examples. Evaluated on HumanEval and MBPP with a 7B model, BatCoder achieved 83.5% and 81.0% pass@1, outperforming strong open-source baselines. Moreover, the framework demonstrates consistent scaling with respect to both training corpus size and model capacity. --- ### Super-résolution non supervisée d'images hyperspectrales de télédétection utilisant un entraînement entièrement synthétique Authors: Xinxin Xu, Yann Gousseau, Christophe Kervazo, Saïd Ladjal Published: 2026-01-30 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2602.02552v1 Hyperspectral single image super-resolution (SISR) aims to enhance spatial resolution while preserving the rich spectral information of hyperspectral images. Most existing methods rely on supervised learning with high-resolution ground truth data, which is often unavailable in practice. To overcome this limitation, we propose an unsupervised learning approach based on synthetic abundance data. The hyperspectral image is first decomposed into endmembers and abundance maps through hyperspectral unmixing. A neural network is then trained to super-resolve these maps using data generated with the dead leaves model, which replicates the statistical properties of real abundances. The final super-resolution hyperspectral image is reconstructed by recombining the super-resolved abundance maps with the endmembers. Experimental results demonstrate the effectiveness of our method and the relevance of synthetic data for training. --- ### EEO-TFV: Escape-Explore Optimizer for Web-Scale Time-Series Forecasting and Vision Analysis Authors: Hua Wang, Jinghao Lu, Fan Zhang Published: 2026-01-30 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2602.02551v1 Transformer-based foundation models have achieved remarkable progress in tasks such as time-series forecasting and image segmentation. However, they frequently suffer from error accumulation in multivariate long-sequence prediction and exhibit vulnerability to out-of-distribution samples in image-related tasks. Furthermore, these challenges become particularly pronounced in large-scale Web data analysis tasks, which typically involve complex temporal patterns and multimodal features. This complexity substantially increases optimization difficulty, rendering models prone to stagnation at saddle points within high-dimensional parameter spaces. To address these issues, we propose a lightweight Transformer architecture in conjunction with a novel Escape-Explore Optimizer (EEO). The optimizer enhances both exploration and generalization while effectively avoiding sharp minima and saddle-point traps. Experimental results show that, in representative Web data scenarios, our method achieves performance on par with state-of-the-art models across 11 time-series benchmark datasets and the Synapse medical image segmentation task. Moreover, it demonstrates superior generalization and stability, thereby validating its potential as a versatile cross-task foundation model for Web-scale data mining and analysis. --- ### HyPAC: Cost-Efficient LLMs-Human Hybrid Annotation with PAC Error Guarantees Authors: Hao Zeng, Huipeng Huang, Xinhao Qu, Jianguo Huang, Bingyi Jing, Hongxin Wei Published: 2026-01-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.02550v1 Data annotation often involves multiple sources with different cost-quality trade-offs, such as fast large language models (LLMs), slow reasoning models, and human experts. In this work, we study the problem of routing inputs to the most cost-efficient annotation source while controlling the labeling error on test instances. We propose \textbf{HyPAC}, a method that adaptively labels inputs to the most cost-efficient annotation source while providing distribution-free guarantees on annotation error. HyPAC calibrates two decision thresholds using importance sampling and upper confidence bounds, partitioning inputs into three regions based on uncertainty and routing each to the appropriate annotation source. We prove that HyPAC achieves the minimum expected cost with a probably approximately correct (PAC) guarantee on the annotation error, free of data distribution and pre-trained models. Experiments on common benchmarks demonstrate the effectiveness of our method, reducing the annotation cost by 78.51\% while tightly controlling the annotation error. --- ### ToolTok: Tool Tokenization for Efficient and Generalizable GUI Agents Authors: Xiaoce Wang, Guibin Zhang, Junzhe Li, Jinzhe Tu, Chun Li, Ming Li Published: 2026-01-30 Categories: cs.LG, cs.AI, cs.CV, cs.MA arXiv: https://arxiv.org/abs/2602.02548v1 Existing GUI agent models relying on coordinate-based one-step visual grounding struggle with generalizing to varying input resolutions and aspect ratios. Alternatives introduce coordinate-free strategies yet suffer from learning under severe data scarcity. To address the limitations, we propose ToolTok, a novel paradigm of multi-step pathfinding for GUI agents, where operations are modeled as a sequence of progressive tool usage. Specifically, we devise tools aligned with human interaction habits and represent each tool using learnable token embeddings. To enable efficient embedding learning under limited supervision, ToolTok introduces a semantic anchoring mechanism that grounds each tool with semantically related concepts as natural inductive bias. To further enable a pre-trained large language model to progressively acquire tool semantics, we construct an easy-to-hard curriculum consisting of three tasks: token definition question-answering, pure text-guided tool selection, and simplified visual pathfinding. Extensive experiments on multiple benchmarks show that ToolTok achieves superior performance among models of comparable scale (4B) and remains competitive with a substantially larger model (235B). Notably, these results are obtained using less than 1% of the training data required by other post-training approaches. In addition, ToolTok demonstrates strong generalization across unseen scenarios. Our training & inference code is open-source at https://github.com/ZephinueCode/ToolTok. --- ### naPINN: Noise-Adaptive Physics-Informed Neural Networks for Recovering Physics from Corrupted Measurement Authors: Hankyeol Kim, Pilsung Kang Published: 2026-01-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.02547v2 Physics-Informed Neural Networks (PINNs) are effective methods for solving inverse problems and discovering governing equations from observational data. However, their performance degrades significantly under complex measurement noise and gross outliers. To address this issue, we propose the Noise-Adaptive Physics-Informed Neural Network (naPINN), which robustly recovers physical solutions from corrupted measurements without prior knowledge of the noise distribution. naPINN embeds an energy-based model into the training loop to learn the latent distribution of prediction residuals. Leveraging the learned energy landscape, a trainable reliability gate adaptively filters data points exhibiting high energy, while a rejection cost regularization prevents trivial solutions where valid data are discarded. We demonstrate the efficacy of naPINN on various benchmark partial differential equations corrupted by non-Gaussian noise and varying rates of outliers. The results show that naPINN significantly outperforms existing robust PINN baselines, successfully isolating outliers and accurately reconstructing the dynamics under severe data corruption. --- ### D$^2$Quant: Accurate Low-bit Post-Training Weight Quantization for LLMs Authors: Xianglong Yan, ChengZhu Bao, Zhiteng Li, Tianao Zhang, Shaoqiu Zhang, Ruobing Xie, Samm Sun, Yulun Zhang Published: 2026-01-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.02546v2 Large language models (LLMs) deliver strong performance, but their high compute and memory costs make deployment difficult in resource-constrained scenarios. Weight-only post-training quantization (PTQ) is appealing, as it reduces memory usage and enables practical speedup without low-bit operators or specialized hardware. However, accuracy often degrades significantly in weight-only PTQ at sub-4-bit precision, and our analysis identifies two main causes: (1) down-projection matrices are a well-known quantization bottleneck, but maintaining their fidelity often requires extra bit-width; (2) weight quantization induces activation deviations, but effective correction strategies remain underexplored. To address these issues, we propose D$^2$Quant, a novel weight-only PTQ framework that improves quantization from both the weight and activation perspectives. On the weight side, we design a Dual-Scale Quantizer (DSQ) tailored to down-projection matrices, with an absorbable scaling factor that significantly improves accuracy without increasing the bit budget. On the activation side, we propose Deviation-Aware Correction (DAC), which incorporates a mean-shift correction within LayerNorm to mitigate quantization-induced activation distribution shifts. Extensive experiments across multiple LLM families and evaluation metrics show that D$^2$Quant delivers superior performance for weight-only PTQ at sub-4-bit precision. The code and models will be available at https://github.com/XIANGLONGYAN/D2Quant. --- ### Beyond Alignment: Expanding Reasoning Capacity via Manifold-Reshaping Policy Optimization Authors: Dayu Wang, Jiaye Yang, Weikang Li, Jiahui Liang, Yang Li Published: 2026-01-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.02545v1 Reinforcement Learning with Verifiable Rewards (RLVR) has demonstrated remarkable success in enhancing the reasoning capabilities of Large Language Models (LLMs). However, recent studies question whether RL genuinely expands reasoning capacity or merely aligns existing latent capabilities, arguing that exploration remains confined within the pre-trained model's low-rank bias manifold. In this work, we challenge this accessibility boundary hypothesis by demonstrating that the latent reasoning space can be fundamentally expanded through targeted geometric interventions. We propose Manifold-Reshaping Policy Optimization (MRPO), a geometric framework designed to fundamentally restructure the inference space of LLMs. MRPO operates in two stages: first, we employ Spectral Orthogonal Exploration (SOE) to eject the policy initialization into the null space of the bias manifold; second, we integrate an Effective Rank regularization term into the policy optimization objective. This approach incentivizes the discovery and maintenance of high-dimensional reasoning trajectories against the entropy-reducing tendency of standard RL. Empirically, our 4B-parameter method achieves state-of-the-art performance on mathematical tasks, significantly outperforming larger models (e.g., Qwen3-32B) and expanding the capability boundary beyond standard GRPO. Our code is available at https://anonymous.4open.science/r/MRPO-D57B/ --- ### SPA-Cache: Singular Proxies for Adaptive Caching in Diffusion Language Models Authors: Wenhao Sun, Rong-Cheng Tu, Yifu Ding, Zhao Jin, Jingyi Liao, Yongcheng Jing, Dacheng Tao Published: 2026-01-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.02544v2 While Diffusion Language Models (DLMs) offer a flexible, arbitrary-order alternative to the autoregressive paradigm, their non-causal nature precludes standard KV caching, forcing costly hidden state recomputation at every decoding step. Existing DLM caching approaches reduce this cost by selective hidden state updates; however, they are still limited by (i) costly token-wise update identification heuristics and (ii) rigid, uniform budget allocation that fails to account for heterogeneous hidden state dynamics. To address these challenges, we present SPA-Cache that jointly optimizes update identification and budget allocation in DLM cache. First, we derive a low-dimensional singular proxy that enables the identification of update-critical tokens in a low-dimensional subspace, substantially reducing the overhead of update identification. Second, we introduce an adaptive strategy that allocates fewer updates to stable layers without degrading generation quality. Together, these contributions significantly improve the efficiency of DLMs, yielding up to an $8\times$ throughput improvement over vanilla decoding and a $2$--$4\times$ speedup over existing caching baselines. --- ### Norm Anchors Make Model Edits Last Authors: Mingda Liu, Zhenghan Zhu, Ze'an Miao, Katsuki Fujisawa Published: 2026-01-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.02543v3 Sequential Locate-and-Edit (L&E) model editing can fail abruptly after many edits. We identify and formalize this failure as a positive norm-feedback loop, in which solved value vectors and edited MLP weights progressively amplify each other, degrading edit quality and eventually collapsing model capabilities. Our analysis shows that this feedback can yield approximately exponential norm growth under standard L&E dynamics, and can remain unresolved by existing increment-level regularizers or update clamps. We propose Norm-Anchor Scaling (NAS), a plug-in stabilizer that breaks this loop by rescaling each solved value vector to an original-model reference norm. Across multiple LLM backbones, datasets, and L&E editors, NAS extends the usable editing horizon by more than 4x and improves long-run editing performance by 72.2% on average, while preserving single-edit efficacy, with only a one-line modification and negligible computational overhead. --- ### Auto-Augmentation Contrastive Learning for Wearable-based Human Activity Recognition Authors: Qingyu Wu, Jianfei Shen, Feiyi Fan, Yang Gu, Chenyang Xu, Yiqiang Chen Published: 2026-01-30 Categories: cs.LG, cs.AI, eess.SP arXiv: https://arxiv.org/abs/2602.02542v1 For low-semantic sensor signals from human activity recognition (HAR), contrastive learning (CL) is essential to implement novel applications or generic models without manual annotation, which is a high-performance self-supervised learning (SSL) method. However, CL relies heavily on data augmentation for pairwise comparisons. Especially for low semantic data in the HAR area, conducting good performance augmentation strategies in pretext tasks still rely on manual attempts lacking generalizability and flexibility. To reduce the augmentation burden, we propose an end-to-end auto-augmentation contrastive learning (AutoCL) method for wearable-based HAR. AutoCL is based on a Siamese network architecture that shares the parameters of the backbone and with a generator embedded to learn auto-augmentation. AutoCL trains the generator based on the representation in the latent space to overcome the disturbances caused by noise and redundant information in raw sensor data. The architecture empirical study indicates the effectiveness of this design. Furthermore, we propose a stop-gradient design and correlation reduction strategy in AutoCL to enhance encoder representation learning. Extensive experiments based on four wide-used HAR datasets demonstrate that the proposed AutoCL method significantly improves recognition accuracy compared with other SOTA methods. --- ### Variational Approach for Job Shop Scheduling Authors: Seung Heon Oh, Jiwon Baek, Hyunjin Oh, Kiyoung Cho, Heechang Yoon, Jong Hun Woo Published: 2026-01-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.00408v3 This paper proposes a novel Variational Graph-to-Scheduler (VG2S) framework for solving the Job Shop Scheduling Problem (JSSP), a critical task in manufacturing that directly impacts operational efficiency and resource utilization. Conventional Deep Reinforcement Learning (DRL) approaches often face challenges such as non-stationarity during training and limited generalization to unseen problem instances because they optimize representation learning and policy execution simultaneously. To address these issues, we introduce variational inference to the JSSP domain for the first time and derive a probabilistic objective based on the Evidence of Lower Bound (ELBO) with maximum entropy reinforcement learning. By mathematically decoupling representation learning from policy optimization, the VG2S framework enables the agent to learn robust structural representations of scheduling instances through a variational graph encoder. This approach significantly enhances training stability and robustness against hyperparameter variations. Extensive experiments demonstrate that the proposed method exhibits superior zero-shot generalization compared with state-of-the-art DRL baselines and traditional dispatching rules, particularly on large-scale and challenging benchmark instances such as DMU and SWV. --- ### Fed-Listing: Federated Label Distribution Inference in Graph Neural Networks Authors: Suprim Nakarmi, Junggab Son, Yue Zhao, Zuobin Xiong Published: 2026-01-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.00407v2 Federated Graph Neural Networks (FedGNNs) facilitate collaborative learning across multiple clients with graph-structured data while preserving user privacy. However, emerging research indicates that within this setting, shared model updates, particularly gradients, can unintentionally leak sensitive information of local users. Numerous privacy inference attacks have been explored in traditional federated learning and extended to graph settings, but the problem of label distribution inference in FedGNNs remains largely underexplored. In this work, we introduce Fed-Listing (Federated Label Distribution Inference in GNNs), a novel gradient-based attack designed to infer the private label statistics of target clients in FedGNNs without access to raw data or node features. Fed-Listing only leverages the final-layer gradients exchanged during training to uncover statistical patterns that reveal class proportions in a stealthy manner. Extensive experiments on four benchmark datasets and three GNN architectures show that Fed-Listing significantly outperforms existing baselines, including random guessing and Decaf, even under challenging non-i.i.d. scenarios. Moreover, existing defense mechanisms can barely reduce the attack performance of Fed-Listing, unless the model's utility is severely degraded. The code implementation and Supplementary materials are available here: https://github.com/suprimnakarmi/Fed-Listing. --- ### RobustDebias: Debiasing Language Models using Distributionally Robust Optimization Authors: Deep Gandhi, Katyani Singh, Nidhi Hegde Published: 2026-01-30 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.00405v1 Pretrained language models have been shown to exhibit biases and social stereotypes. Prior work on debiasing these models has largely focused on modifying embedding spaces during pretraining, which is not scalable for large models. Fine-tuning pretrained models on task-specific datasets can both degrade model performance and amplify biases present in the fine-tuning data. We address bias amplification during fine-tuning rather than costly pretraining, focusing on BERT models due to their widespread use in language understanding tasks. While Empirical Risk Minimization effectively optimizes downstream performance, it often amplifies social biases during fine-tuning. To counter this, we propose \textit{RobustDebias}, a novel mechanism which adapts Distributionally Robust Optimization (DRO) to debias language models during fine-tuning. Our approach debiases models across multiple demographics during MLM fine-tuning and generalizes to any dataset or task. Extensive experiments on various language models show significant bias mitigation with minimal performance impact. --- ### DROGO: Default Representation Objective via Graph Optimization in Reinforcement Learning Authors: Hon Tik Tse, Marlos C. Machado Published: 2026-01-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.00403v1 In computational reinforcement learning, the default representation (DR) and its principal eigenvector have been shown to be effective for a wide variety of applications, including reward shaping, count-based exploration, option discovery, and transfer. However, in prior investigations, the eigenvectors of the DR were computed by first approximating the DR matrix, and then performing an eigendecomposition. This procedure is computationally expensive and does not scale to high-dimensional spaces. In this paper, we derive an objective for directly approximating the principal eigenvector of the DR with a neural network. We empirically demonstrate the effectiveness of the objective in a number of environments, and apply the learned eigenvectors for reward shaping. --- ### A Conditional Companion: Lived Experiences of People with Mental Health Disorders Using LLMs Authors: Aditya Kumar Purohit, Hendrik Heuer Published: 2026-01-30 Categories: cs.HC, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2602.00402v2 Large Language Models (LLMs) are increasingly used for mental health support, yet little is known about how people with mental health challenges engage with them, how they evaluate their usefulness, and what design opportunities they envision. We conducted 20 semi-structured interviews with people in the UK who live with mental health conditions and have used LLMs for mental health support. Through reflexive thematic analysis, we found that participants engaged with LLMs in conditional and situational ways: for immediacy, the desire for non-judgement, self-paced disclosure, cognitive reframing, and relational engagement. Simultaneously, participants articulated clear boundaries informed by prior therapeutic experience: LLMs were effective for mild-to-moderate distress but inadequate for crises, trauma, and complex social-emotional situations. We contribute empirical insights into the lived use of LLMs for mental health, highlight boundary-setting as central to their safe role, and propose design and governance directions for embedding them responsibly within care ecosystem. --- ### ZEST: Zero-shot Embodied Skill Transfer for Athletic Robot Control Authors: Jean Pierre Sleiman, He Li, Alphonsus Adu-Bredu, Robin Deits, Arun Kumar, Kevin Bergamin, Mohak Bhardwaj, Scott Biddlestone, Nicola Burger, Matthew A. Estrada, Francesco Iacobelli, Twan Koolen et al. Published: 2026-01-30 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2602.00401v1 Achieving robust, human-like whole-body control on humanoid robots for agile, contact-rich behaviors remains a central challenge, demanding heavy per-skill engineering and a brittle process of tuning controllers. We introduce ZEST (Zero-shot Embodied Skill Transfer), a streamlined motion-imitation framework that trains policies via reinforcement learning from diverse sources -- high-fidelity motion capture, noisy monocular video, and non-physics-constrained animation -- and deploys them to hardware zero-shot. ZEST generalizes across behaviors and platforms while avoiding contact labels, reference or observation windows, state estimators, and extensive reward shaping. Its training pipeline combines adaptive sampling, which focuses training on difficult motion segments, and an automatic curriculum using a model-based assistive wrench, together enabling dynamic, long-horizon maneuvers. We further provide a procedure for selecting joint-level gains from approximate analytical armature values for closed-chain actuators, along with a refined model of actuators. Trained entirely in simulation with moderate domain randomization, ZEST demonstrates remarkable generality. On Boston Dynamics' Atlas humanoid, ZEST learns dynamic, multi-contact skills (e.g., army crawl, breakdancing) from motion capture. It transfers expressive dance and scene-interaction skills, such as box-climbing, directly from videos to Atlas and the Unitree G1. Furthermore, it extends across morphologies to the Spot quadruped, enabling acrobatics, such as a continuous backflip, through animation. Together, these results demonstrate robust zero-shot deployment across heterogeneous data sources and embodiments, establishing ZEST as a scalable interface between biological movements and their robotic counterparts. --- ### KEPO: Knowledge-Enhanced Preference Optimization for Multimodal Reasoning with Applications to Medical VQA Authors: Fan Yang, Rui Meng, Trudi Di Qi, Ali Ezzati, Yuxin Wen Published: 2026-01-30 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.00400v2 Reinforcement learning (RL) has emerged as a promising paradigm for inducing explicit reasoning behaviors in large language and vision-language models. However, reasoning-oriented RL post-training remains fundamentally challenging due to sparse trajectory-level rewards, leading to ambiguous credit assignment and severe exploration failures that can trap the policy in a ``learning cliff.'' Recent on-policy distillation methods introduce dense teacher supervision to stabilize optimization, but apply it uniformly across all generated trajectories. We argue that such uniform distillation is ill-suited for reasoning-intensive tasks, as low-quality on-policy trajectories often originate from early logical errors, and distillation under flawed contexts injects noisy and misaligned gradients. To address these challenges, we propose Knowledge-Enhanced Preference Optimization (KEPO), a unified post-training framework that integrates: (i) a quality-gated on-policy distillation objective that selectively applies dense teacher guidance only to high-quality trajectories, and (ii) a knowledge-enhanced exploration strategy that leverages hints learned from a teacher model to rejectively sample reward-positive on-policy trajectories for RL, thereby mitigating exploration collapse. Evaluated on a challenging medical visual question answering benchmark under single-source generalization, KEPO demonstrates improved training stability, more coherent reasoning behaviors, and superior out-of-distribution performance over reinforcement learning and on-policy distillation baselines. --- ### Reinforcement Learning for Control Systems with Time Delays: A Comprehensive Survey Authors: Armando Alves Neto Published: 2026-01-30 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2602.00399v1 In the last decade, Reinforcement Learning (RL) has achieved remarkable success in the control and decision-making of complex dynamical systems. However, most RL algorithms rely on the Markov Decision Process assumption, which is violated in practical cyber-physical systems affected by sensing delays, actuation latencies, and communication constraints. Such time delays introduce memory effects that can significantly degrade performance and compromise stability, particularly in networked and multi-agent environments. This paper presents a comprehensive survey of RL methods designed to address time delays in control systems. We first formalize the main classes of delays and analyze their impact on the Markov property. We then systematically categorize existing approaches into five major families: state augmentation and history-based representations, recurrent policies with learned memory, predictor-based and model-aware methods, robust and domain-randomized training strategies, and safe RL frameworks with explicit constraint handling. For each family, we discuss underlying principles, practical advantages, and inherent limitations. A comparative analysis highlights key trade-offs among these approaches and provides practical guidelines for selecting suitable methods under different delay characteristics and safety requirements. Finally, we identify open challenges and promising research directions, including stability certification, large-delay learning, multi-agent communication co-design, and standardized benchmarking. This survey aims to serve as a unified reference for researchers and practitioners developing reliable RL-based controllers in delay-affected cyber-physical systems. --- ### MemoryLLM: Plug-n-Play Interpretable Feed-Forward Memory for Transformers Authors: Ajay Jaiswal, Lauren Hannah, Han-Byul Kim, Duc Hoang, Arnav Kundu, Mehrdad Farajtabar, Minsik Cho Published: 2026-01-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.00398v1 Understanding how transformer components operate in LLMs is important, as it is at the core of recent technological advances in artificial intelligence. In this work, we revisit the challenges associated with interpretability of feed-forward modules (FFNs) and propose MemoryLLM, which aims to decouple FFNs from self-attention and enables us to study the decoupled FFNs as context-free token-wise neural retrieval memory. In detail, we investigate how input tokens access memory locations within FFN parameters and the importance of FFN memory across different downstream tasks. MemoryLLM achieves context-free FFNs by training them in isolation from self-attention directly using the token embeddings. This approach allows FFNs to be pre-computed as token-wise lookups (ToLs), enabling on-demand transfer between VRAM and storage, additionally enhancing inference efficiency. We also introduce Flex-MemoryLLM, positioning it between a conventional transformer design and MemoryLLM. This architecture bridges the performance gap caused by training FFNs with context-free token-wise embeddings. --- ### Fast Forward: Accelerating LLM Prefill with Predictive FFN Sparsity Authors: Aayush Gautam, Mukul Gagrani, Junyoung Park, Mingu Lee, Chiris Lott, Narasimha Reddy Published: 2026-01-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.00397v1 The prefill stage of large language model (LLM) inference is a key computational bottleneck for long-context workloads. At short-to-moderate context lengths (1K--16K tokens), Feed-Forward Networks (FFNs) dominate this cost, accounting for most of the total FLOPs. Existing FFN sparsification methods, designed for autoregressive decoding, fail to exploit the prefill stage's parallelism and often degrade accuracy. To address this, we introduce FastForward, a predictive sparsity framework that accelerates LLM prefill through block-wise, context-aware FFN sparsity. FastForward combines (1) a lightweight expert predictor to select high-importance neurons per block, (2) an error compensation network to correct sparsity-induced errors, and (3) a layer-wise sparsity scheduler to allocate compute based on token-mixing importance. Across LLaMA and Qwen models up to 8B parameters, FastForward delivers up to 1.45$\times$ compute-bound speedup at 50% FFN sparsity with $<$ 6% accuracy loss compared to the dense baseline on LongBench, substantially reducing Time-to-First-Token (TTFT) for efficient, long-context LLM inference on constrained hardware. --- ### 3DGS$^2$-TR: Scalable Second-Order Trust-Region Method for 3D Gaussian Splatting Authors: Roger Hsiao, Yuchen Fang, Xiangru Huang, Ruilong Li, Hesam Rabeti, Zan Gojcic, Javad Lavaei, James Demmel, Sophia Shao Published: 2026-01-30 Categories: cs.CV, cs.LG, math.OC arXiv: https://arxiv.org/abs/2602.00395v1 We propose 3DGS$^2$-TR,a second-order optimizer for accelerating the scene training problem in 3D Gaussian Splatting (3DGS). Unlike existing second-order approaches that rely on explicit or dense curvature representations, such as 3DGS-LM (Höllein et al., 2025) or 3DGS2 (Lan et al., 2025), our method approximates curvature using only the diagonal of the Hessian matrix, efficiently via Hutchinson's method. Our approach is fully matrix-free and has the same complexity as ADAM (Kingma, 2024), $O(n)$ in both computation and memory costs. To ensure stable optimization in the presence of strong nonlinearity in the 3DGS rasterization process, we introduce a parameter-wise trust-region technique based on the squared Hellinger distance, regularizing updates to Gaussian parameters. Under identical parameter initialization and without densification, 3DGS$^2$-TR is able to achieve better reconstruction quality on standard datasets, using 50% fewer training iterations compared to ADAM, while incurring less than 1GB of peak GPU memory overhead (17% more than ADAM and 85% less than 3DGS-LM), enabling scalability to very large scenes and potentially to distributed training settings. --- ### Modeling Art Evaluations from Comparative Judgments: A Deep Learning Approach to Predicting Aesthetic Preferences Authors: Manoj Reddy Bethi, Sai Rupa Jhade, Pravallika Yaganti, Monoshiz Mahbub Khan, Zhe Yu Published: 2026-01-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.00394v1 Modeling human aesthetic judgments in visual art presents significant challenges due to individual preference variability and the high cost of obtaining labeled data. To reduce cost of acquiring such labels, we propose to apply a comparative learning framework based on pairwise preference assessments rather than direct ratings. This approach leverages the Law of Comparative Judgment, which posits that relative choices exhibit less cognitive burden and greater cognitive consistency than direct scoring. We extract deep convolutional features from painting images using ResNet-50 and develop both a deep neural network regression model and a dual-branch pairwise comparison model. We explored four research questions: (RQ1) How does the proposed deep neural network regression model with CNN features compare to the baseline linear regression model using hand-crafted features? (RQ2) How does pairwise comparative learning compare to regression-based prediction when lacking access to direct rating values? (RQ3) Can we predict individual rater preferences through within-rater and cross-rater analysis? (RQ4) What is the annotation cost trade-off between direct ratings and comparative judgments in terms of human time and effort? Our results show that the deep regression model substantially outperforms the baseline, achieving up to $328\%$ improvement in $R^2$. The comparative model approaches regression performance despite having no access to direct rating values, validating the practical utility of pairwise comparisons. However, predicting individual preferences remains challenging, with both within-rater and cross-rater performance significantly lower than average rating prediction. Human subject experiments reveal that comparative judgments require $60\%$ less annotation time per item, demonstrating superior annotation efficiency for large-scale preference modeling. --- ### Brazilian Portuguese Image Captioning with Transformers: A Study on Cross-Native-Translated Dataset Authors: Gabriel Bromonschenkel, Alessandro L. Koerich, Thiago M. Paixão, Hilário Tomaz Alves de Oliveira Published: 2026-01-30 Categories: cs.CV, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.00393v1 Image captioning (IC) refers to the automatic generation of natural language descriptions for images, with applications ranging from social media content generation to assisting individuals with visual impairments. While most research has been focused on English-based models, low-resource languages such as Brazilian Portuguese face significant challenges due to the lack of specialized datasets and models. Several studies create datasets by automatically translating existing ones to mitigate resource scarcity. This work addresses this gap by proposing a cross-native-translated evaluation of Transformer-based vision and language models for Brazilian Portuguese IC. We use a version of Flickr30K comprised of captions manually created by native Brazilian Portuguese speakers and compare it to a version with captions automatically translated from English to Portuguese. The experiments include a cross-context approach, where models trained on one dataset are tested on the other to assess the translation impact. Additionally, we incorporate attention maps for model inference interpretation and use the CLIP-Score metric to evaluate the image-description alignment. Our findings show that Swin-DistilBERTimbau consistently outperforms other models, demonstrating strong generalization across datasets. ViTucano, a Brazilian Portuguese pre-trained VLM, surpasses larger multilingual models (GPT-4o, LLaMa 3.2 Vision) in traditional text-based evaluation metrics, while GPT-4 models achieve the highest CLIP-Score, highlighting improved image-text alignment. Attention analysis reveals systematic biases, including gender misclassification, object enumeration errors, and spatial inconsistencies. The datasets and the models generated and analyzed during the current study are available in: https://github.com/laicsiifes/transformer-caption-ptbr. --- ### Localized, High-resolution Geographic Representations with Slepian Functions Authors: Arjun Rao, Ruth Crasto, Tessa Ooms, David Rolnick, Konstantin Klemmer, Marc Rußwurm Published: 2026-01-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.00392v2 Geographic data is fundamentally local. Disease outbreaks cluster in population centers, ecological patterns emerge along coastlines, and economic activity concentrates within country borders. Machine learning models that encode geographic location, however, distribute representational capacity uniformly across the globe, struggling at the fine-grained resolutions that localized applications require. We propose a geographic location encoder built from spherical Slepian functions that concentrate representational capacity inside a region-of-interest and scale to high resolutions without extensive computational demands. For settings requiring global context, we present a hybrid Slepian-Spherical Harmonic encoder that efficiently bridges the tradeoff between local-global performance, while retaining desirable properties such as pole-safety and spherical-surface-distance preservation. Across five tasks spanning classification, regression, and image-augmented prediction, Slepian encodings outperform baselines and retain performance advantages across a wide range of neural network architectures. --- ### Robust automatic brain vessel segmentation in 3D CTA scans using dynamic 4D-CTA data Authors: Alberto Mario Ceballos-Arroyo, Shrikanth M. Yadav, Chu-Hsuan Lin, Jisoo Kim, Geoffrey S. Young, Lei Qin, Huaizu Jiang Published: 2026-01-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.00391v3 In this study, we develop a novel methodology for annotating the brain vasculature using dynamic 4D-CTA head scans. By using multiple time points from dynamic CTA acquisitions, we subtract bone and soft tissue to enhance the visualization of arteries and veins, reducing the effort required to obtain manual annotations of brain vessels. We then train deep learning models on our ground truth annotations by using the same segmentation for multiple phases from the dynamic 4D-CTA collection, effectively enlarging our dataset by 4 to 5 times and inducing robustness to contrast phases. In total, our dataset comprises 110 training images from 25 patients and 165 test images from 14 patients. In comparison with two similarly-sized datasets for CTA-based brain vessel segmentation, a nnUNet model trained on our dataset can achieve significantly better segmentations across all vascular regions, with an average mDC of 0.846 for arteries and 0.957 for veins in the TopBrain dataset. Furthermore, metrics such as average directed Hausdorff distance (adHD) and topology sensitivity (tSens) reflected similar trends: using our dataset resulted in low error margins (adHD of 0.304 mm for arteries and 0.078 for veins) and high sensitivity (tSens of 0.877 for arteries and 0.974 for veins), indicating excellent accuracy in capturing vessel morphology. Our code and model weights are available online at https://github.com/alceballosa/robust-vessel-segmentation --- ### Safer by Diffusion, Broken by Context: Diffusion LLM's Safety Blessing and Its Failure Mode Authors: Zeyuan He, Yupeng Chen, Lang Lin, Yihan Wang, Shenxu Chang, Eric Sommerlade, Philip Torr, Junchi Yu, Adel Bibi, Jialin Yu Published: 2026-01-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.00388v2 Diffusion large language models (D-LLMs) offer an alternative to autoregressive LLMs (AR-LLMs) and have demonstrated advantages in generation efficiency. Beyond the utility benefits, we argue that D-LLMs exhibit a previously underexplored safety blessing: their diffusion-style generation confers intrinsic robustness against jailbreak attacks originally designed for AR-LLMs. In this work, we provide an initial analysis of the underlying mechanism, showing that the diffusion trajectory induces a stepwise reduction effect that progressively suppresses unsafe generations. This robustness, however, is not absolute. Following this analysis, we highlight a simple yet effective failure mode, context nesting, in which harmful requests are embedded within structured benign contexts. Empirically, we show that this simple black-box strategy bypasses D-LLMs' safety blessing, achieving state-of-the-art attack success rates across models and benchmarks. Notably, it enables the first successful jailbreak of Gemini Diffusion to our knowledge, exposing a critical vulnerability in proprietary D-LLMs. Together, our results characterize both the origins and the limits of D-LLMs' safety blessing, constituting an early-stage red-teaming of D-LLMs. --- ### Singular Bayesian Neural Networks Authors: Mame Diarra Toure, David A. Stephens Published: 2026-01-30 Categories: stat.ML, cs.LG, stat.AP arXiv: https://arxiv.org/abs/2602.00387v4 Bayesian neural networks promise calibrated uncertainty but require $O(mn)$ parameters for standard mean-field Gaussian posteriors. We argue this cost is often unnecessary, particularly when weight matrices exhibit fast singular value decay. By parameterizing weights as $W = AB^{\top}$ with $A \in \mathbb{R}^{m \times r}$, $B \in \mathbb{R}^{n \times r}$, we induce a posterior that is \emph{singular} with respect to the Lebesgue measure, concentrating on the rank-$r$ manifold. This singularity captures structured weight correlations through shared latent factors, geometrically distinct from mean-field's independence assumption. We derive PAC-Bayes generalization bounds whose complexity term scales as $\sqrt{r(m+n)}$ instead of $\sqrt{m n}$, and prove loss bounds that decompose the error into optimization and rank-induced bias using the Eckart-Young-Mirsky theorem. We further adapt recent Gaussian complexity bounds for low-rank deterministic networks to Bayesian predictive means. Empirically, across MLPs, LSTMs, and Transformers on standard benchmarks, our method achieves competitive predictive performance while using up to $33\times$ fewer parameters than 5-member Deep Ensembles. It substantially improves OOD detection and often improves calibration relative to mean-field and perturbation baselines, while Deep Ensembles can still be stronger on in-distribution likelihood-based metrics. --- ### Generalized Inverses of Matrix Products: From Fundamental Subspaces to Randomized Decompositions Authors: Michał P. Karpowicz, Gilbert Strang Published: 2026-01-30 Categories: math.NA, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.00386v1 We investigate the Moore-Penrose pseudoinverse and generalized inverse of a matrix product $A=CR$ to establish a unifying framework for generalized and randomized matrix inverses. This analysis is rooted in first principles, focusing on the geometry of the four fundamental subspaces. We examine: (1) the reverse order law, $A^+ = R^+C^+$, which holds when $C$ has independent columns and $R$ has independent rows, (2) the universally correct formula, $A^+ = (C^+CR)^+(CRR^+)^+$, providing a geometric interpretation of the mappings between the involved subspaces, (3) a new generalized randomized formula, $A^+_p = (P^TA)^+P^TAQ(AQ)^+$, which gives $A^+_p = A^+$ if and only if the sketching matrices $P$ and $Q$ preserve the rank of $A$, i.e., $\mathrm{rank}(P^TA) = \mathrm{rank}(AQ) = \mathrm{rank}(A)$. The framework is extended to generalized $\{1,2\}$-inverses and specialized forms, revealing the underlying structure of established randomized linear algebra algorithms, including randomized SVD, the Nyström approximation, and CUR decomposition. We demonstrate applications in sparse sensor placement and effective resistance estimation. For the latter, we provide a rigorous quantitative analysis of an approximation scheme, establishing that it always underestimates the true resistance and deriving a worst-case spectral bound on the error of resistance differences. --- ### Deep Learning-Based Object Detection for Autonomous Vehicles: A Comparative Study of One-Stage and Two-Stage Detectors on Basic Traffic Objects Authors: Bsher Karbouj, Adam Michael Altenbuchner, Joerg Krueger Published: 2026-01-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.00385v1 Object detection is a crucial component in autonomous vehicle systems. It enables the vehicle to perceive and understand its environment by identifying and locating various objects around it. By utilizing advanced imaging and deep learning techniques, autonomous vehicle systems can rapidly and accurately identify objects based on their features. Different deep learning methods vary in their ability to accurately detect and classify objects in autonomous vehicle systems. Selecting the appropriate method significantly impacts system performance, robustness, and efficiency in real-world driving scenarios. While several generic deep learning architectures like YOLO, SSD, and Faster R-CNN have been proposed, guidance on their suitability for specific autonomous driving applications is often limited. The choice of method affects detection accuracy, processing speed, environmental robustness, sensor integration, scalability, and edge case handling. This study provides a comprehensive experimental analysis comparing two prominent object detection models: YOLOv5 (a one-stage detector) and Faster R-CNN (a two-stage detector). Their performance is evaluated on a diverse dataset combining real and synthetic images, considering various metrics including mean Average Precision (mAP), recall, and inference speed. The findings reveal that YOLOv5 demonstrates superior performance in terms of mAP, recall, and training efficiency, particularly as dataset size and image resolution increase. However, Faster R-CNN shows advantages in detecting small, distant objects and performs well in challenging lighting conditions. The models' behavior is also analyzed under different confidence thresholds and in various real-world scenarios, providing insights into their applicability for autonomous driving systems. --- ### RePaint-Enhanced Conditional Diffusion Model for Parametric Engineering Designs under Performance and Parameter Constraints Authors: Ke Wang, Nguyen Gia Hien Vu, Yifan Tang, Mostafa Rahmani Dehaghani, G. Gary Wang Published: 2026-01-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.00384v1 This paper presents a RePaint-enhanced framework that integrates a pre-trained performance-guided denoising diffusion probabilistic model (DDPM) for performance- and parameter-constraint engineering design generation. The proposed method enables the generation of missing design components based on a partial reference design while satisfying performance constraints, without retraining the underlying model. By applying mask-based resampling during inference process, RePaint allows efficient and controllable repainting of partial designs under both performance and parameter constraints, which is not supported by conventional DDPM-base methods. The framework is evaluated on two representative design problems, parametric ship hull design and airfoil design, demonstrating its ability to generate novel designs with expected performance based on a partial reference design. Results show that the method achieves accuracy comparable to or better than pre-trained models while enabling controlled novelty through fixing partial designs. Overall, the proposed approach provides an efficient, training-free solution for parameter-constraint-aware generative design in engineering applications. --- ### Modeling Image-Caption Rating from Comparative Judgments Authors: Kezia Minni, Qiang Zhang, Monoshiz Mahbub Khan, Zhe Yu Published: 2026-01-30 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2602.00381v2 Image caption rating is becoming increasingly important because computer-generated captions are used extensively for descriptive annotation. However, rating the accuracy of captions in describing images is time-consuming and subjective in nature. In contrast, it is often easier for people to compare (between two pairs) which image-caption pair better matches each other. In this study, we propose a machine learning framework that models such comparative judgments instead of direct ratings. The model can then be applied to rank unseen image-caption pairs in the same way as a regression model trained on direct ratings. Inspired by a state-of-the-art regression approach, we extracted visual and text features using a pre-trained ViLBERT model and tweaked the learning parameters of the baseline model to improve the model performance. This new regression model (with Kendall's $τ_c=0.812$) outperformed the baseline model (with Kendall's $τ_c=0.758$) on the VICR dataset. The same model structure was applied to the comparative learning framework. Trained on comparative judgments (image-caption pair A better matches each other than image-caption pair B), the comparative learning model achieved a performance similar (with Kendall's $τ_c=0.804$) to that of the regression model. In addition, a small-scale human subject study was conducted to compare the cost and quality of direct ratings, pairwise comparisons, and same-image comparisons. The results showed that comparative judgments yielded faster results and greater agreement among human annotators than direct ratings. These results suggest that collecting comparative judgments instead of direct ratings as training data labels is promising for lower annotation costs and greater consistency. The model trained on such comparative judgments can perform as well as the model trained on direct ratings. --- ### Clause-Internal or Clause-External? Testing Turkish Reflexive Binding in Adapted versus Chain of Thought Large Language Models Authors: Sercan Karakaş Published: 2026-01-30 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.00380v2 This study evaluates whether state-of-the-art large language models capture the binding relations of Turkish reflexive pronouns. We construct a balanced evaluation set of 100 Turkish sentences that systematically pit local against non-local antecedents for the reflexives kendi and kendisi. We compare two contrasting systems: an OpenAI chain-of-thought model optimized for multi-step reasoning and Trendyol-LLM-7B-base-v0.1, a LLaMA 2 derived model extensively fine-tuned on Turkish data. Antecedent choice is assessed using a combined paradigm that integrates sentence-level perplexity with a forced-choice comparison between minimally differing continuations. Overall, Trendyol-LLM favors local bindings in approximately 70 percent of trials, exhibiting a robust locality bias consistent with a preference for structurally proximate antecedents. By contrast, the OpenAI model (o1 Mini) distributes its choices nearly evenly between local and long-distance readings, suggesting weaker or less consistent sensitivity to locality in this binding configuration. Taken together, these results reveal a marked contrast in binding behavior across the two systems and motivate closer analysis of how model architecture, training data, and inference-time reasoning strategies shape the representation of Turkish anaphoric dependencies. --- ### Parametrization of subgrid scales in long-term simulations of the shallow-water equations using machine learning and convex limiting Authors: Md Amran Hossan Mojamder, Zhihang Xu, Min Wang, Ilya Timofeyev Published: 2026-01-30 Categories: physics.flu-dyn, cs.LG, physics.ao-ph, physics.comp-ph arXiv: https://arxiv.org/abs/2602.00378v2 We present a method for parametrizing sub-grid processes in the Shallow Water equations. We define coarse variables and local spatial averages and use a feed-forward neural network to learn sub-grid fluxes. Our method results in a local parametrization that uses a four-point computational stencil, which has several advantages over globally coupled parametrizations. We demonstrate numerically that our method improves energy balance in long-term turbulent simulations and also accurately reproduces individual solutions. The long-term simulations refer to numerical studies where a fluid flow is simulated over a duration long enough to reach a statistical steady state. The neural network parametrization can be easily combined with flux limiting to reduce oscillations near shocks. More importantly, our method provides reliable parametrizations, even in dynamical regimes that are not included in the training data. --- ### DecompressionLM: Deterministic, Diagnostic, and Zero-Shot Concept Graph Extraction from Language Models Authors: Zhaochen Hong, Jiaxuan You Published: 2026-01-30 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.00377v3 Existing knowledge probing methods rely on pre-defined queries, limiting extraction to known concepts. We introduce DecompressionLM, a stateless framework for zero-shot concept graph extraction that discovers what language models encode without pre-specified queries or shared cross-sequence state. Our method targets three limitations of common decoding-based probing approaches: (i) cross-sequence coupling that concentrates probability mass on high-frequency prefixes, (ii) competitive decoding effects that suppress long-tail concepts, and (iii) scalability constraints arising from sequential exploration. Using Van der Corput low-discrepancy sequences with arithmetic decoding, DecompressionLM enables deterministic, embarrassingly parallel generation without shared state across sequences. Across two model families and five quantization variants, we find that activation-aware quantization (AWQ-4bit) expands concept coverage by 30-170%, while uniform quantization (GPTQ-Int4) induces 71-86% coverage collapse - divergent behaviors not reliably reflected by explanation-level perplexity. Corpus-based verification further reveals a 19.6-point hallucination gap between top- and bottom-ranked MMLU-Pro Law models. DecompressionLM establishes concept coverage as a complementary evaluation dimension for assessing knowledge breadth and factual grounding in compressed models intended for deployment. Code is available at https://github.com/ulab-uiuc/decompressionlm. --- ### MATRIX: A Multimodal Benchmark and Post-Training Framework for Materials Science Authors: Delia McGrath, Curtis Chong, Rohil Kulkarni, Gerbrand Ceder, Adeesh Kolluru Published: 2026-01-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.00376v1 Scientific reasoning in materials science requires integrating multimodal experimental evidence with underlying physical theory. Existing benchmarks make it difficult to assess whether incorporating visual experimental data during post-training improves mechanism-grounded explanation reasoning beyond text-only supervision. We introduce MATRIX, a multimodal benchmark for materials science reasoning that evaluates foundational theory, research-level reasoning, and the interpretation of real experimental artifacts across multiple characterization modalities. Using MATRIX as a controlled diagnostic, we isolate the effect of visual grounding by comparing post-training on structured materials science text alone with post-training that incorporates paired experimental images. Despite using relatively small amounts of multimodal data, visual supervision improves experimental interpretation by 10-25% and yields 5-16% gains on text-only scientific reasoning tasks. Our results demonstrate that these improvements rely on correct image-text alignment during post-training, highlighting cross-modal representational transfer. We also observe consistent improvements on ScienceQA and PubMedQA, demonstrating that the benefits of structured multimodal post-training extend beyond materials science. The MATRIX dataset is available at https://huggingface.co/datasets/radical-ai/MATRIX and the model at https://huggingface.co/radical-ai/MATRIX-PT. --- ### Post-Training Probability Manifold Correction via Structured SVD Pruning and Self-Referential Distillation Authors: Aaron R. Flouro, Shawn P. Chadwick Published: 2026-01-30 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2602.00372v1 Large language models are expensive to deploy. We introduce Sparse Knowledge Distillation (SparseKD), a post-training method that compresses transformer models by combining structured SVD pruning with self-referential knowledge distillation. The key insight is simple: instead of using an external teacher, the model teaches itself by matching its own probability distribution from before compression. This self-referential setup enables surprisingly strong quality recovery after aggressive pruning. Our experiments reveal an unexpected finding: self-referential distillation alone, applied post-training under an identical objective and fixed calibration dataset, improves model quality by 39% relative to the original converged checkpoint. When combined with structured pruning, SparseKD achieves 15-65% parameter reduction with acceptable quality trade-offs. Kernel profiling shows that speedups arise entirely from reduced dense matrix multiplication in feed-forward layers while attention remains unchanged, making this approach complementary to attention optimizations. We validate across two model families (0.6B and 3.8B parameters) with multi-seed experiments confirming high reproducibility. SparseKD requires no external super-teacher, no architectural changes, and no custom inference kernels, making it immediately deployable with existing infrastructure. --- ### POET: Protocol Optimization via Eligibility Tuning Authors: Trisha Das, Katherine Kero, Dorinda Schumann, Tracy Ohrt, Sanjit Singh Batra, Gregory D Lyng, Robert E. Tillman Published: 2026-01-30 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.00370v1 Eligibility criteria (EC) are essential for clinical trial design, yet drafting them remains a time-intensive and cognitively demanding task for clinicians. Existing automated approaches often fall at two extremes either requiring highly structured inputs, such as predefined entities to generate specific criteria, or relying on end-to-end systems that produce full eligibility criteria from minimal input such as trial descriptions limiting their practical utility. In this work, we propose a guided generation framework that introduces interpretable semantic axes, such as Demographics, Laboratory Parameters, and Behavioral Factors, to steer EC generation. These axes, derived using large language models, offer a middle ground between specificity and usability, enabling clinicians to guide generation without specifying exact entities. In addition, we present a reusable rubric-based evaluation framework that assesses generated criteria along clinically meaningful dimensions. Our results show that our guided generation approach consistently outperforms unguided generation in both automatic, rubric-based and clinician evaluations, offering a practical and interpretable solution for AI-assisted trial design. --- ### Quantum Generator Kernels Authors: Philipp Altmann, Maximilian Mansky, Maximilian Zorn, Jonas Stein, Claudia Linnhoff-Popien Published: 2026-01-30 Categories: cs.LG, quant-ph arXiv: https://arxiv.org/abs/2602.00361v1 Quantum kernel methods offer significant theoretical benefits by rendering classically inseparable features separable in quantum space. Yet, the practical application of Quantum Machine Learning (QML), currently constrained by the limitations of Noisy Intermediate-Scale Quantum (NISQ) hardware, necessitates effective strategies to compress and embed large-scale real-world data like images into the constrained capacities of existing quantum devices or simulators. To this end, we propose Quantum Generator Kernels (QGKs), a generator-based approach to quantum kernels, comprising a set of Variational Generator Groups (VGGs) that merge universal generators into a parameterizable operator, ensuring scalable coverage of the available quantum space. Thereby, we address shortcomings of current leading strategies employing hybrid architectures, which might prevent exploiting quantum computing's full potential due to fixed intermediate embedding processes. To optimize the kernel alignment to the target domain, we train a weight vector to parameterize the projection of the VGGs in the current data context. Our empirical results demonstrate superior projection and classification capabilities of the QGK compared to state-of-the-art quantum and classical kernel approaches and show its potential to serve as a versatile framework for various QML applications. --- ### Leveraging Textual-Cues for Enhancing Multimodal Sentiment Analysis by Object Recognition Authors: Sumana Biswas, Karen Young, Josephine Griffith Published: 2026-01-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.00360v1 Multimodal sentiment analysis, which includes both image and text data, presents several challenges due to the dissimilarities in the modalities of text and image, the ambiguity of sentiment, and the complexities of contextual meaning. In this work, we experiment with finding the sentiments of image and text data, individually and in combination, on two datasets. Part of the approach introduces the novel `Textual-Cues for Enhancing Multimodal Sentiment Analysis' (TEMSA) based on object recognition methods to address the difficulties in multimodal sentiment analysis. Specifically, we extract the names of all objects detected in an image and combine them with associated text; we call this combination of text and image data TEMS. Our results demonstrate that only TEMS improves the results when considering all the object names for the overall sentiment of multimodal data compared to individual analysis. This research contributes to advancing multimodal sentiment analysis and offers insights into the efficacy of TEMSA in combining image and text data for multimodal sentiment analysis. --- ### Position: Agentic Evolution is the Path to Evolving LLMs Authors: Minhua Lin, Hanqing Lu, Zhan Shi, Bing He, Rui Mao, Zhiwei Zhang, Zongyu Wu, Xianfeng Tang, Hui Liu, Zhenwei Dai, Xiang Zhang, Suhang Wang et al. Published: 2026-01-30 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.00359v2 As Large Language Models (LLMs) move from curated training sets into open-ended real-world environments, a fundamental limitation emerges: static training cannot keep pace with continual deployment environment change. Scaling training-time and inference-time compute improves static capability but does not close this train-deploy gap. We argue that addressing this limitation requires a new scaling axis-evolution. Existing deployment-time adaptation methods, whether parametric fine-tuning or heuristic memory accumulation, lack the strategic agency needed to diagnose failures and produce durable improvements. Our position is that agentic evolution represents the inevitable future of LLM adaptation, elevating evolution itself from a fixed pipeline to an autonomous evolver agent. We instantiate this vision in a general framework, A-Evolve, which treats deployment-time improvement as a deliberate, goal-directed optimization process over persistent system state. We further propose the evolution-scaling hypothesis: the capacity for adaptation scales with the compute allocated to evolution, positioning agentic evolution as a scalable path toward sustained, open-ended adaptation in the real world. Our code is publicly available at https://github.com/A-EVO-Lab/a-evolve. --- ### Planning with Language and Generative Models: Toward General Reward-Guided Wireless Network Design Authors: Chenyang Yuan, Xiaoyuan Cheng Published: 2026-01-30 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.00357v1 Intelligent access point (AP) deployment remains challenging in next-generation wireless networks due to complex indoor geometries and signal propagation. We firstly benchmark general-purpose large language models (LLMs) as agentic optimizers for AP planning and find that, despite strong wireless domain knowledge, their dependence on external verifiers results in high computational costs and limited scalability. Motivated by these limitations, we study generative inference models guided by a unified reward function capturing core AP deployment objectives across diverse floorplans. We show that diffusion samplers consistently outperform alternative generative approaches. The diffusion process progressively improves sampling by smoothing and sharpening the reward landscape, rather than relying on iterative refinement, which is effective for non-convex and fragmented objectives. Finally, we introduce a large-scale real-world dataset for indoor AP deployment, requiring over $50k$ CPU hours to train general reward functions, and evaluate in- and out-of-distribution generalization and robustness. Our results suggest that diffusion-based generative inference with a unified reward function provides a scalable and domain-agnostic foundation for indoor AP deployment planning. --- ### MHDash: An Online Platform for Benchmarking Mental Health-Aware AI Assistants Authors: Yihe Zhang, Cheyenne N Mohawk, Kaiying Han, Vijay Srinivas Tida, Manyu Li, Xiali Hei Published: 2026-01-30 Categories: cs.AI, cs.HC arXiv: https://arxiv.org/abs/2602.00353v1 Large language models (LLMs) are increasingly applied in mental health support systems, where reliable recognition of high-risk states such as suicidal ideation and self-harm is safety-critical. However, existing evaluations primarily rely on aggregate performance metrics, which often obscure risk-specific failure modes and provide limited insight into model behavior in realistic, multi-turn interactions. We present MHDash, an open-source platform designed to support the development, evaluation, and auditing of AI systems for mental health applications. MHDash integrates data collection, structured annotation, multi-turn dialogue generation, and baseline evaluation into a unified pipeline. The platform supports annotations across multiple dimensions, including Concern Type, Risk Level, and Dialogue Intent, enabling fine-grained and risk-aware analysis. Our results reveal several key findings: (i) simple baselines and advanced LLM APIs exhibit comparable overall accuracy yet diverge significantly on high-risk cases; (ii) some LLMs maintain consistent ordinal severity ranking while failing absolute risk classification, whereas others achieve reasonable aggregate scores but suffer from high false negative rates on severe categories; and (iii) performance gaps are amplified in multi-turn dialogues, where risk signals emerge gradually. These observations demonstrate that conventional benchmarks are insufficient for safety-critical mental health settings. By releasing MHDash as an open platform, we aim to promote reproducible research, transparent evaluation, and safety-aligned development of AI systems for mental health support. --- ### DETOUR: An Interactive Benchmark for Dual-Agent Search and Reasoning Authors: Li Siyan, Darshan Deshpande, Anand Kannappan, Rebecca Qian Published: 2026-01-30 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.00352v1 When recalling information in conversation, people often arrive at the recollection after multiple turns. However, existing benchmarks for evaluating agent capabilities in such tip-of-the-tongue search processes are restricted to single-turn settings. To more realistically simulate tip-of-the-tongue search, we introduce Dual-agent based Evaluation Through Obscure Under-specified Retrieval (DETOUR), a dual-agent evaluation benchmark containing 1,011 prompts. The benchmark design involves a Primary Agent, which is the subject of evaluation, tasked with identifying the recollected entity through querying a Memory Agent that is held consistent across evaluations. Our results indicate that current state-of-the-art models still struggle with our benchmark, only achieving 36% accuracy when evaluated on all modalities (text, image, audio, and video), highlighting the importance of enhancing capabilities in underspecified scenarios. --- ### ReLAPSe: Reinforcement-Learning-trained Adversarial Prompt Search for Erased concepts in unlearned diffusion models Authors: Ignacy Kolton, Kacper Marzol, Paweł Batorski, Marcin Mazur, Paul Swoboda, Przemysław Spurek Published: 2026-01-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.00350v1 Machine unlearning is a key defense mechanism for removing unauthorized concepts from text-to-image diffusion models, yet recent evidence shows that latent visual information often persists after unlearning. Existing adversarial approaches for exploiting this leakage are constrained by fundamental limitations: optimization-based methods are computationally expensive due to per-instance iterative search. At the same time, reasoning-based and heuristic techniques lack direct feedback from the target model's latent visual representations. To address these challenges, we introduce ReLAPSe, a policy-based adversarial framework that reformulates concept restoration as a reinforcement learning problem. ReLAPSe trains an agent using Reinforcement Learning with Verifiable Rewards (RLVR), leveraging the diffusion model's noise prediction loss as a model-intrinsic and verifiable feedback signal. This closed-loop design directly aligns textual prompt manipulation with latent visual residuals, enabling the agent to learn transferable restoration strategies rather than optimizing isolated prompts. By pioneering the shift from per-instance optimization to global policy learning, ReLAPSe achieves efficient, near-real-time recovery of fine-grained identities and styles across multiple state-of-the-art unlearning methods, providing a scalable tool for rigorous red-teaming of unlearned diffusion models. Some experimental evaluations involve sensitive visual concepts, such as nudity. Code is available at https://github.com/gmum/ReLaPSe --- ### MASC: Metal-Aware Sampling and Correction via Reinforcement Learning for Accelerated MRI Authors: Zhengyi Lu, Ming Lu, Chongyu Qu, Junchao Zhu, Junlin Guo, Marilyn Lionts, Yanfan Zhu, Yuechen Yang, Tianyuan Yao, Jayasai Rajagopal, Bennett Allan Landman, Xiao Wang et al. Published: 2026-01-30 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.00348v1 Metal implants in MRI cause severe artifacts that degrade image quality and hinder clinical diagnosis. Traditional approaches address metal artifact reduction (MAR) and accelerated MRI acquisition as separate problems. We propose MASC, a unified reinforcement learning framework that jointly optimizes metal-aware k-space sampling and artifact correction for accelerated MRI. To enable supervised training, we construct a paired MRI dataset using physics-based simulation, generating k-space data and reconstructions for phantoms with and without metal implants. This paired dataset provides simulated 3D MRI scans with and without metal implants, where each metal-corrupted sample has an exactly matched clean reference, enabling direct supervision for both artifact reduction and acquisition policy learning. We formulate active MRI acquisition as a sequential decision-making problem, where an artifact-aware Proximal Policy Optimization (PPO) agent learns to select k-space phase-encoding lines under a limited acquisition budget. The agent operates on undersampled reconstructions processed through a U-Net-based MAR network, learning patterns that maximize reconstruction quality. We further propose an end-to-end training scheme where the acquisition policy learns to select k-space lines that best support artifact removal while the MAR network simultaneously adapts to the resulting undersampling patterns. Experiments demonstrate that MASC's learned policies outperform conventional sampling strategies, and end-to-end training improves performance compared to using a frozen pre-trained MAR network, validating the benefit of joint optimization. Cross-dataset experiments on FastMRI with physics-based artifact simulation further confirm generalization to realistic clinical MRI data. The code and models of MASC have been made publicly available: https://github.com/hrlblab/masc ---