# Research Papers — 2026-03 · bundle page 206 of 243 > Full-text markdown of papers #10251–10300 of 12123 published in 2026-03, newest first. AI-readable mirrors of the full 12123-paper 2026-03 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-03/page/206 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/205.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/207.md) ### WebChain: A Large-Scale Human-Annotated Dataset of Real-World Web Interaction Traces Authors: Sicheng Fan, Rui Wan, Yifei Leng, Gaoning Liang, Li Ling, Yanyi Shang, Dehan Kong Published: 2026-03-05 Categories: cs.AI, cs.CV arXiv: https://arxiv.org/abs/2603.05295v3 We introduce WebChain, the largest open-source dataset of human-annotated trajectories on real-world websites, designed to accelerate reproducible research in web agents. It contains 31,725 trajectories and 318k steps, featuring a core Triple Alignment of visual, structural, and action data to provide rich, multi-modal supervision. The data is collected via a scalable pipeline that ensures coverage of complex, high-value tasks often missed by synthetic methods. Leveraging this dataset, we propose a Dual Mid-Training recipe that decouples spatial grounding from planning, achieving state-of-the-art performance on our proposed WebChainBench and other public GUI benchmarks. Our work provides the data and insights necessary to build and rigorously evaluate the next generation of scalable web agents. --- ### STRUCTUREDAGENT: Planning with AND/OR Trees for Long-Horizon Web Tasks Authors: ELita Lobo, Xu Chen, Jingjing Meng, Nan Xi, Yang Jiao, Chirag Agarwal, Yair Zick, Yan Gao Published: 2026-03-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.05294v3 Existing LLM-based web agents struggle on complex, long-horizon tasks due to limited in-context memory, weak planning abilities, and greedy behaviors that lead to premature termination. To address these challenges, we propose \SA{}, a hierarchical planning framework that interleaves planning and execution via dynamic $\ANDOR$ trees. The framework separates structural planning from LLM-based reasoning, enabling principled error recovery through node repair, systematic exploration of alternatives via OR nodes, and modular plans that can facilitate human intervention. On WebArena (630 tasks), \SA{} achieves a $\sim$53\% success rate vs.\ $\sim$46\% for AgentOccam, and on complex multi-constraint Amazon shopping tasks, gains reach 10\% over the strongest baseline. --- ### STRUCTUREDAGENT: Planning with AND/OR Trees for Long-Horizon Web Tasks Authors: ELita Lobo, Xu Chen, Jingjing Meng, Nan Xi, Yang Jiao, Chirag Agarwal, Yair Zick, Yan Gao Published: 2026-03-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.05294v2 Recent advances in large language models (LLMs) have enabled agentic systems for sequential decision-making. Such agents must perceive their environment, reason across multiple time steps, and take actions that optimize long-term objectives. However, existing web agents struggle on complex, long-horizon tasks due to limited in-context memory for tracking history, weak planning abilities, and greedy behaviors that lead to premature termination. To address these challenges, we propose STRUCTUREDAGENT, a hierarchical planning framework with two core components: (1) an online hierarchical planner that uses dynamic AND/OR trees for efficient search and (2) a structured memory module that tracks and maintains candidate solutions to improve constraint satisfaction in information-seeking tasks. The framework also produces interpretable hierarchical plans, enabling easier debugging and facilitating human intervention when needed. Our results on WebVoyager, WebArena, and custom shopping benchmarks show that STRUCTUREDAGENT improves performance on long-horizon web-browsing tasks compared to standard LLM-based agents. --- ### Knowledge Divergence and the Value of Debate for Scalable Oversight Authors: Robin Young Published: 2026-03-05 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2603.05293v1 AI safety via debate and reinforcement learning from AI feedback (RLAIF) are both proposed methods for scalable oversight of advanced AI systems, yet no formal framework relates them or characterizes when debate offers an advantage. We analyze this by parameterizing debate's value through the geometry of knowledge divergence between debating models. Using principal angles between models' representation subspaces, we prove that the debate advantage admits an exact closed form. When models share identical training corpora, debate reduces to RLAIF-like where a single-agent method recovers the same optimum. When models possess divergent knowledge, debate advantage scales with a phase transition from quadratic regime (debate offers negligible benefit) to linear regime (debate is essential). We classify three regimes of knowledge divergence (shared, one-sided, and compositional) and provide existence results showing that debate can achieve outcomes inaccessible to either model alone, alongside a negative result showing that sufficiently strong adversarial incentives cause coordination failure in the compositional regime, with a sharp threshold separating effective from ineffective debate. We offer the first formal connection between debate and RLAIF, a geometric foundation for understanding when adversarial oversight protocols are justified, and connection to the problem of eliciting latent knowledge across models with complementary information. --- ### X-RAY: Mapping LLM Reasoning Capability via Formalized and Calibrated Probes Authors: Tianxi Gao, Yufan Cai, Yusi Yuan, Jin Song Dong Published: 2026-03-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.05290v2 Large language models (LLMs) achieve promising performance, yet their ability to reason remains poorly understood. Existing evaluations largely emphasize task-level accuracy, often conflating pattern matching with reasoning capability. We present X-RAY, an explainable reasoning analysis system that maps the LLM reasoning capability using calibrated, formally verified probes. We model reasoning capability as a function of extractable \textit{structure}, operationalized through formal properties such as constraint interaction, reasoning depth, and solution-space geometry. X-Ray generates probes via formal tools with controlled structural variations, enabling precise isolation of incremental structural information through formal calibration and verification. We evaluate state-of-the-art LLMs on problems ranging from junior-level to advanced in mathematics, physics, and chemistry. Our analysis reveals a systematic asymmetry in LLM reasoning: models are relatively robust to constraint refinement, where additional conditions shrink an existing solution space, but degrade sharply under solution-space restructuring, where modifications alter the underlying structural form of the solution manifold. Moreover, calibrated formal probes differentiate models that appear indistinguishable on standard benchmarks and reveal failure modes that are structurally interpretable rather than opaque. Beyond evaluation, our framework is contamination-free and supports the training and testing of reasoning models. --- ### Bayesian Supervised Causal Clustering Authors: Luwei Wang, Nazir Lone, Sohan Seth Published: 2026-03-05 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2603.05288v1 Finding patient subgroups with similar characteristics is crucial for personalized decision-making in various disciplines such as healthcare and policy evaluation. While most existing approaches rely on unsupervised clustering methods, there is a growing trend toward using supervised clustering methods that identify operationalizable subgroups in the context of a specific outcome of interest. We propose Bayesian Supervised Causal Clustering (BSCC), with treatment effect as outcome to guide the clustering process. BSCC identifies homogenous subgroups of individuals who are similar in their covariate profiles as well as their treatment effects. We evaluate BSCC on simulated datasets as well as real-world dataset from the third International Stroke Trial to assess the practical usefulness of the framework. --- ### Layer by layer, module by module: Choose both for optimal OOD probing of ViT Authors: Ambroise Odonnat, Vasilii Feofanov, Laetitia Chapel, Romain Tavenard, Ievgen Redko Published: 2026-03-05 Categories: cs.CV, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2603.05280v1 Recent studies have observed that intermediate layers of foundation models often yield more discriminative representations than the final layer. While initially attributed to autoregressive pretraining, this phenomenon has also been identified in models trained via supervised and discriminative self-supervised objectives. In this paper, we conduct a comprehensive study to analyze the behavior of intermediate layers in pretrained vision transformers. Through extensive linear probing experiments across a diverse set of image classification benchmarks, we find that distribution shift between pretraining and downstream data is the primary cause of performance degradation in deeper layers. Furthermore, we perform a fine-grained analysis at the module level. Our findings reveal that standard probing of transformer block outputs is suboptimal; instead, probing the activation within the feedforward network yields the best performance under significant distribution shift, whereas the normalized output of the multi-head self-attention module is optimal when the shift is weak. --- ### Whispering to a Blackbox: Bootstrapping Frozen OCR with Visual Prompts Authors: Samandar Samandarov, Nazirjon Ismoiljonov, Abdullah Sattorov, Temirlan Sabyrbayev Published: 2026-03-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.05276v1 In the landscape of modern machine learning, frozen pre-trained models provide stability and efficiency but often underperform on specific tasks due to mismatched data distributions. This paper introduces the Whisperer, a novel visual prompting framework that learns diffusion-based preprocessors to adapt inputs in pixel space, effectively "whispering" enhancements to frozen downstream models like EasyOCR. By framing the process as behavioral cloning of stochastically discovered improvement policies, our method achieves an 8% absolute (10.6% relative) reduction in Character Error Rate (CER) on a challenging dataset of 300k degraded synthetic text images, surpassing hand-engineered baselines such as CLAHE. The key innovation is a four-stage training curriculum that uses behavioral cloning to amplify "lucky" improvements discovered through the stochastic exploration of a partially trained diffusion model. This approach is highly sample-efficient and avoids the pitfalls of traditional reinforcement learning. Crucially, we frame this not as naive reinforcement learning, but as behavioral cloning of an exploration policy: we stochastically sample intermediate diffusion outputs, select those that improve CER by chance, and then train the model to reproduce them. This bootstrapping curriculum (4 stages over 60 GPU-hours) amplifies random successes into a systematic strategy. In summary, by whispering to the frozen OCR through its inputs, we improve an imperfect classifier without touching its weights. --- ### SarcasmMiner: A Dual-Track Post-Training Framework for Robust Audio-Visual Sarcasm Reasoning Authors: Zhu Li, Yongjian Chen, Huiyuan Lai, Xiyuan Gao, Shekhar Nayak, Matt Coler Published: 2026-03-05 Categories: cs.MM, cs.CL, cs.SD arXiv: https://arxiv.org/abs/2603.05275v1 Multimodal sarcasm detection requires resolving pragmatic incongruity across textual, acoustic, and visual cues through cross-modal reasoning. To enable robust sarcasm reasoning with foundation models, we propose SarcasmMiner, a reinforcement learning based post-training framework that resists hallucination in multimodal reasoning. We reformulate sarcasm detection as structured reasoning and adopt a dual-track distillation strategy: high-quality teacher trajectories initialize the student model, while the full set of trajectories trains a generative reward model (GenRM) to evaluate reasoning quality. The student is optimized with group relative policy optimization (GRPO) using decoupled rewards for accuracy and reasoning quality. On MUStARD++, SarcasmMiner increases F1 from 59.83% (zero-shot), 68.23% (supervised finetuning) to 70.22%. These findings suggest that reasoning-aware reward modeling enhances both performance and multimodal grounding. --- ### Oral to Web: Digitizing 'Zero Resource'Languages of Bangladesh Authors: Mohammad Mamun Or Rashid Published: 2026-03-05 Categories: cs.CL, cs.HC arXiv: https://arxiv.org/abs/2603.05272v2 We present the Multilingual Cloud Corpus, the first national-scale, parallel, multimodal linguistic dataset of Bangladesh's ethnic and indigenous languages. Despite being home to approximately 40 minority languages spanning four language families, Bangladesh has lacked a systematic, cross-family digital corpus for these predominantly oral, computationally "zero resource" varieties, 14 of which are classified as endangered. Our corpus comprises 85792 structured textual entries, each containing a Bengali stimulus text, an English translation, and an IPA transcription, together with approximately 107 hours of transcribed audio recordings, covering 42 language varieties from the Tibeto-Burman, Indo-European, Austro-Asiatic, and Dravidian families, plus two genetically unclassified languages. The data were collected through systematic fieldwork over 90 days across nine districts of Bangladesh, involving 16 data collectors, 77 speakers, and 43 validators, following a predefined elicitation template of 2224 unique items organized at three levels of linguistic granularity: isolated lexical items (475 words across 22 semantic domains), grammatical constructions (887 sentences across 21 categories including verbal conjugation paradigms), and directed speech (862 prompts across 46 conversational scenarios). Post-field processing included IPA transcription by 10 linguists with independent adjudication by 6 reviewers. The complete dataset is publicly accessible through the Multilingual Cloud platform (multiling.cloud), providing searchable access to annotated audio and textual data for all documented varieties. We describe the corpus design, fieldwork methodology, dataset structure, and per-language coverage, and discuss implications for endangered language documentation, low-resource NLP, and digital preservation in linguistically diverse developing countries. --- ### Visual-Informed Speech Enhancement Using Attention-Based Beamforming Authors: Chihyun Liu, Jiaxuan Fan, Mingtung Sun, Michael Anthony, Mingsian R. Bai, Yu Tsao Published: 2026-03-05 Categories: eess.AS, cs.AI arXiv: https://arxiv.org/abs/2603.05270v1 Recent studies have demonstrated that incorporating auxiliary information, such as speaker voiceprint or visual cues, can substantially improve Speech Enhancement (SE) performance. However, single-channel methods often yield suboptimal results in low signal-to-noise ratio (SNR) conditions, when there is high reverberation, or in complex scenarios involving dynamic speakers, overlapping speech, or non-stationary noise. To address these issues, we propose a novel Visual-Informed Neural Beamforming Network (VI-NBFNet), which integrates microphone array signal processing and deep neural networks (DNNs) using multimodal input features. The proposed network leverages a pretrained visual speech recognition model to extract lip movements as input features, which serve for voice activity detection (VAD) and target speaker identification. The system is intended to handle both static and moving speakers by introducing a supervised end-to-end beamforming framework equipped with an attention mechanism. The experimental results demonstrated that the proposed audiovisual system has achieved better SE performance and robustness for both stationary and dynamic speaker scenarios, compared to several baseline methods. --- ### Beyond Word Error Rate: Auditing the Diversity Tax in Speech Recognition through Dataset Cartography Authors: Ting-Hui Cheng, Line H. Clemmensen, Sneha Das Published: 2026-03-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.05267v1 Automatic speech recognition (ASR) systems are predominantly evaluated using the Word Error Rate (WER). However, raw token-level metrics fail to capture semantic fidelity and routinely obscures the `diversity tax', the disproportionate burden on marginalized and atypical speaker due to systematic recognition failures. In this paper, we explore the limitations of relying solely on lexical counts by systematically evaluating a broader class of non-linear and semantic metrics. To enable rigorous model auditing, we introduce the sample difficulty index (SDI), a novel metric that quantifies how intrinsic demographic and acoustic factors drive model failure. By mapping SDI on data cartography, we demonstrate that metrics EmbER and SemDist expose hidden systemic biases and inter-model disagreements that WER ignores. Finally, our findings are the first steps towards a robust audit framework for prospective safety analysis, empowering developers to audit and mitigate ASR disparities prior to deployment. --- ### A Behaviour-Aware Federated Forecasting Framework for Distributed Stand-Alone Wind Turbines Authors: Bowen Li, Xiufeng Liu, Maria Sinziiana Astefanoaei Published: 2026-03-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.05263v1 Accurate short-term wind power forecasting is essential for grid dispatch and market operations, yet centralising turbine data raises privacy, cost, and heterogeneity concerns. We propose a two-stage federated learning framework that first clusters turbines by long-term behavioural statistics using Double Roulette Selection (DRS) initialisation with recursive Auto-split refinement, and then trains cluster-specific LSTM models via FedAvg. Experiments on 400 stand-alone turbines in Denmark show that DRS-auto discovers behaviourally coherent groups and achieves competitive forecasting accuracy while preserving data locality. Behaviour-aware grouping consistently outperforms geographic partitioning and matches strong k-means++ baselines, suggesting a practical privacy-friendly solution for heterogeneous distributed turbine fleets. --- ### VietJobs: A Vietnamese Job Advertisement Dataset Authors: Hieu Pham Dinh, Hung Nguyen Huy, Mo El-Haj Published: 2026-03-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.05262v1 VietJobs is the first large-scale, publicly available corpus of Vietnamese job advertisements, comprising 48,092 postings and over 15 million words collected from all 34 provinces and municipalities across Vietnam. The dataset provides extensive linguistic and structured information, including job titles, categories, salaries, skills, and employment conditions, covering 16 occupational domains and multiple employment types (full-time, part-time, and internship). Designed to support research in natural language processing and labour market analytics, VietJobs captures substantial linguistic, regional, and socio-economic diversity. We benchmark several generative large language models (LLMs) on two core tasks: job category classification and salary estimation. Instruction-tuned models such as Qwen2.5-7B-Instruct and Llama-SEA-LION-v3-8B-IT demonstrate notable gains under few-shot and fine-tuned settings, while highlighting challenges in multilingual and Vietnamese-specific modelling for structured labour market prediction. VietJobs establishes a new benchmark for Vietnamese NLP and offers a valuable foundation for future research on recruitment language, socio-economic representation, and AI-driven labour market analysis. All code and resources are available at: https://github.com/VinNLP/VietJobs. --- ### Wiki-R1: Incentivizing Multimodal Reasoning for Knowledge-based VQA via Data and Sampling Curriculum Authors: Shan Ning, Longtian Qiu, Xuming He Published: 2026-03-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.05256v2 Knowledge-Based Visual Question Answering (KB-VQA) requires models to answer questions about an image by integrating external knowledge, posing significant challenges due to noisy retrieval and the structured, encyclopedic nature of the knowledge base. These characteristics create a distributional gap from pretrained multimodal large language models (MLLMs), making effective reasoning and domain adaptation difficult in the post-training stage. In this work, we propose \textit{Wiki-R1}, a data-generation-based curriculum reinforcement learning framework that systematically incentivizes reasoning in MLLMs for KB-VQA. Wiki-R1 constructs a sequence of training distributions aligned with the model's evolving capability, bridging the gap from pretraining to the KB-VQA target distribution. We introduce \textit{controllable curriculum data generation}, which manipulates the retriever to produce samples at desired difficulty levels, and a \textit{curriculum sampling strategy} that selects informative samples likely to yield non-zero advantages during RL updates. Sample difficulty is estimated using observed rewards and propagated to unobserved samples to guide learning. Experiments on two KB-VQA benchmarks, Encyclopedic VQA and InfoSeek, demonstrate that Wiki-R1 achieves new state-of-the-art results, improving accuracy from 35.5\% to 37.1\% on Encyclopedic VQA and from 40.1\% to 44.1\% on InfoSeek. The project page is available at https://artanic30.github.io/project_pages/WikiR1/. --- ### CATNet: Collaborative Alignment and Transformation Network for Cooperative Perception Authors: Gong Chen, Chaokun Zhang, Tao Tang, Pengcheng Lv, Feng Li, Xin Xie Published: 2026-03-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.05255v1 Cooperative perception significantly enhances scene understanding by integrating complementary information from diverse agents. However, existing research often overlooks critical challenges inherent in real-world multi-source data integration, specifically high temporal latency and multi-source noise. To address these practical limitations, we propose Collaborative Alignment and Transformation Network (CATNet), an adaptive compensation framework that resolves temporal latency and noise interference in multi-agent systems. Our key innovations can be summarized in three aspects. First, we introduce a Spatio-Temporal Recurrent Synchronization (STSync) that aligns asynchronous feature streams via adjacent-frame differential modeling, establishing a temporal-spatially unified representation space. Second, we design a Dual-Branch Wavelet Enhanced Denoiser (WTDen) that suppresses global noise and reconstructs localized feature distortions within aligned representations. Third, we construct an Adaptive Feature Selector (AdpSel) that dynamically focuses on critical perceptual features for robust fusion. Extensive experiments on multiple datasets demonstrate that CATNet consistently outperforms existing methods under complex traffic conditions, proving its superior robustness and adaptability. --- ### ICHOR: A Robust Representation Learning Approach for ASL CBF Maps with Self-Supervised Masked Autoencoders Authors: Xavier Beltran-Urbano, Yiran Li, Xinglin Zeng, Katie R. Jobson, Manuel Taso, Christopher A. Brown, David A. Wolk, Corey T. McMillan, Ilya M. Nashrallah, Paul A. Yushkevich, Ze Wang, John A. Detre et al. Published: 2026-03-05 Categories: eess.IV, cs.CV, physics.med-ph arXiv: https://arxiv.org/abs/2603.05247v1 Arterial spin labeling (ASL) perfusion MRI allows direct quantification of regional cerebral blood flow (CBF) without exogenous contrast, enabling noninvasive measurements that can be repeated without constraints imposed by contrast injection. ASL is increasingly acquired in research studies and clinical MRI protocols. Building on successes in structural imaging, recent efforts have implemented deep learning based methods to improve image quality, enable automated quality control, and derive robust quantitative and predictive biomarkers with ASL derived CBF. However, progress has been limited by variable image quality, substantial inter-site, vendor and protocol differences, and limited availability of labeled datasets needed to train models that generalize across cohorts. To address these challenges, we introduce ICHOR, a self supervised pre-training approach for ASL CBF maps that learns transferable representations using 3D masked autoencoders. ICHOR is pretrained via masked image modeling using a Vision Transformer backbone and can be used as a general-purpose encoder for downstream ASL tasks. For pre-training, we curated one of the largest ASL datasets to date, comprising 11,405 ASL CBF scans from 14 studies spanning multiple sites and acquisition protocols. We evaluated the pre-trained ICHOR encoder on three downstream diagnostic classification tasks and one ASL CBF map quality prediction regression task. Across all evaluations, ICHOR outperformed existing neuroimaging self-supervised pre-training methods adapted to ASL. Pre-trained weights and code will be made publicly available. --- ### GCAgent: Enhancing Group Chat Communication through Dialogue Agents System Authors: Zijie Meng, Zheyong Xie, Zheyu Ye, Chonggang Lu, Zuozhu Liu, Zihan Niu, Yao Hu, Shaosheng Cao Published: 2026-03-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.05240v1 As a key form in online social platforms, group chat is a popular space for interest exchange or problem-solving, but its effectiveness is often hindered by inactivity and management challenges. While recent large language models (LLMs) have powered impressive one-to-one conversational agents, their seamlessly integration into multi-participant conversations remains unexplored. To address this gap, we introduce GCAgent, an LLM-driven system for enhancing group chats communication with both entertainment- and utility-oriented dialogue agents. The system comprises three tightly integrated modules: Agent Builder, which customizes agents to align with users' interests; Dialogue Manager, which coordinates dialogue states and manage agent invocations; and Interface Plugins, which reduce interaction barriers by three distinct tools. Through extensive experiment, GCAgent achieved an average score of 4.68 across various criteria and was preferred in 51.04\% of cases compared to its base model. Additionally, in real-world deployments over 350 days, it increased message volume by 28.80\%, significantly improving group activity and engagement. Overall, this work presents a practical blueprint for extending LLM-based dialogue agent from one-party chats to multi-party group scenarios. --- ### Reclaiming Lost Text Layers for Source-Free Cross-Domain Few-Shot Learning Authors: Zhenyu Zhang, Guangyao Chen, Yixiong Zou, Yuhua Li, Ruixuan Li Published: 2026-03-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.05235v1 Source-Free Cross-Domain Few-Shot Learning (SF-CDFSL) focuses on fine-tuning with limited training data from target domains (e.g., medical or satellite images), where CLIP has recently shown promising results due to its generalizability to downstream tasks. Current works indicate CLIP's text encoder is more suitable for cross-domain tasks, however, we find that \textbf{removing certain middle layers of the text encoder can effectively improve performance in SF-CDFSL}, which we call the Lost Layers. In this paper, we delve into this phenomenon for a deeper understanding. We discover that instead of being harmful for the SF-CDFSL task, the information in these layers is actually beneficial, but visual gaps prevent this useful information from being fully utilized, making these layers seem redundant. Based on this understanding, unlike current works that simply remove these layers, we propose a method to teachs the model to \textbf{re-utilize} information in these lost layers at both the layer and encoder levels, guiding the re-learning of the visual branch under domain shifts. Our approach effectively addresses the issue of underutilized information in the text encoder. Extensive experiments across various settings, backbones (CLIP, SigLip, PE-Core), and tasks (4 CDFSL datasets and 10 Meta-dataset datasets) demonstrate the effectiveness of our method. Code is available at https://github.com/zhenyuZ-HUST/CVPR26-VtT. --- ### Recursive Inference Machines for Neural Reasoning Authors: Mieszko Komisarczyk, Saurabh Mathur, Maurice Kraus, Sriraam Natarajan, Kristian Kersting Published: 2026-03-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.05234v1 Neural reasoners such as Tiny Recursive Models (TRMs) solve complex problems by combining neural backbones with specialized inference schemes. Such inference schemes have been a central component of stochastic reasoning systems, where inference rules are applied to a stochastic model to derive answers to complex queries. In this work, we bridge these two paradigms by introducing Recursive Inference Machines (RIMs), a neural reasoning framework that explicitly incorporates recursive inference mechanisms inspired by classical inference engines. We show that TRMs can be expressed as an instance of RIMs, allowing us to extend them through a reweighting component, yielding better performance on challenging reasoning benchmarks, including ARC-AGI-1, ARC-AGI-2, and Sudoku Extreme. Furthermore, we show that RIMs can be used to improve reasoning on other tasks, such as the classification of tabular data, outperforming TabPFNs. --- ### SlideSparse: Fast and Flexible (2N-2):2N Structured Sparsity Authors: Hanyong Shao, Yingbo Hao, Ting Song, Yan Xia, Di Zhang, Shaohan Huang, Xun Wu, Songchen Xu, Le Xu, Li Dong, Zewen Chi, Yi Zou et al. Published: 2026-03-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.05232v1 NVIDIA's 2:4 Sparse Tensor Cores deliver 2x throughput but demand strict 50% pruning -- a ratio that collapses LLM reasoning accuracy (Qwen3: 54% to 15%). Milder $(2N-2):2N$ patterns (e.g., 6:8, 25% pruning) preserve accuracy yet receive no hardware support, falling back to dense execution without any benefit from sparsity. We present SlideSparse, the first system to unlock Sparse Tensor Core acceleration for the $(2N-2):2N$ model family on commodity GPUs. Our Sliding Window Decomposition reconstructs any $(2N-2):2N$ weight block into $N-1$ overlapping 2:4-compliant windows without any accuracy loss; Activation Lifting fuses the corresponding activation rearrangement into per-token quantization at near-zero cost. Integrated into vLLM, SlideSparse is evaluated across various GPUs (A100, H100, B200, RTX 4090, RTX 5080, DGX-spark), precisions (FP4, INT8, FP8, BF16, FP16), and model families (Llama, Qwen, BitNet). On compute-bound workloads, the measured speedup ratio (1.33x) approaches the theoretical upper-bound $N/(N-1)=4/3$ at 6:8 weight sparsity in Qwen2.5-7B, establishing $(2N-2):2N$ as a practical path to accuracy-preserving LLM acceleration. Code available at https://github.com/bcacdwk/vllmbench. --- ### Boosting ASR Robustness via Test-Time Reinforcement Learning with Audio-Text Semantic Rewards Authors: Linghan Fang, Tianxin Xie, Li Liu Published: 2026-03-05 Categories: cs.SD, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.05231v1 Recently, Automatic Speech Recognition (ASR) systems (e.g., Whisper) have achieved remarkable accuracy improvements but remain highly sensitive to real-world unseen data (data with large distribution shifts), including noisy environments and diverse accents. To address this issue, test-time adaptation (TTA) has shown great potential in improving the model adaptability at inference time without ground-truth labels, and existing TTA methods often rely on pseudo-labeling or entropy minimization. However, by treating model confidence as a learning signal, these methods may reinforce high-confidence errors, leading to confirmation bias that undermines adaptation. To overcome these limitations, we present ASR-TRA, a novel Test-time Reinforcement Adaptation framework inspired by causal intervention. More precisely, our method introduces a learnable decoder prompt and utilizes temperature-controlled stochastic decoding to generate diverse transcription candidates. These are scored by a reward model that measures audio-text semantic alignment, and the resulting feedback is used to update both model and prompt parameters via reinforcement learning. Comprehensive experiments on LibriSpeech with synthetic noise and L2 Arctic accented English datasets demonstrate that our method achieves higher accuracy while maintaining lower latency than existing TTA baselines. Ablation studies further confirm the effectiveness of combining audio and language-based rewards, highlighting our method's enhanced stability and interpretability. Overall, our approach provides a practical and robust solution for deploying ASR systems in challenging real-world conditions. --- ### Digital Twin Driven Textile Classification and Foreign Object Recognition in Automated Sorting Systems Authors: Serkan Ergun, Tobias Mitterer, Hubert Zangl Published: 2026-03-05 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2603.05230v2 The increasing demand for sustainable textile recycling requires robust automation solutions capable of handling deformable garments and detecting foreign objects in cluttered environments. This work presents a digital twin driven robotic sorting system that integrates grasp prediction, multi modal perception, and semantic reasoning for real world textile classification. A dual arm robotic cell equipped with RGBD sensing, capacitive tactile feedback, and collision-aware motion planning autonomously separates garments from an unsorted basket, transfers them to an inspection zone, and classifies them using state of the art Visual Language Models (VLMs). We benchmark nine VLM s from five model families on a dataset of 223 inspection scenarios comprising shirts, socks, trousers, underwear, foreign objects (including garments outside of the aforementioned classes), and empty scenes. The evaluation assesses per class accuracy, hallucination behavior, and computational performance under practical hardware constraints. Results show that the Qwen model family achieves the highest overall accuracy (up to 87.9 %), with strong foreign object detection performance, while lighter models such as Gemma3 offer competitive speed accuracy trade offs for edge deployment. A digital twin combined with MoveIt enables collision aware path planning and integrates segmented 3D point clouds of inspected garments into the virtual environment for improved manipulation reliability. The presented system demonstrates the feasibility of combining semantic VLM reasoning with conventional grasp detection and digital twin technology for scalable, autonomous textile sorting in realistic industrial settings. --- ### Not All Trust is the Same: Effects of Decision Workflow and Explanations in Human-AI Decision Making Authors: Laura Spillner, Rachel Ringe, Robert Porzel, Rainer Malaka Published: 2026-03-05 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2603.05229v1 A central challenge in AI-assisted decision making is achieving warranted, well-calibrated trust. Both overtrust (accepting incorrect AI recommendations) and undertrust (rejecting correct advice) should be prevented. Prior studies differ in the design of the decision workflow - whether users see the AI suggestion immediately (1-step setup) or have to submit a first decision beforehand (2-step setup) -, and in how trust is measured - through self-reports or as behavioral trust, that is, reliance. We examined the effects and interactions of (a) the type of decision workflow, (b) the presence of explanations, and (c) users' domain knowledge and prior AI experience. We compared reported trust, reliance (agreement rate and switch rate), and overreliance. Results showed no evidence that a 2-step setup reduces overreliance. The decision workflow also did not directly affect self-reported trust, but there was a crossover interaction effect with domain knowledge and explanations, suggesting that the effects of explanations alone may not generalize across workflow setups. Finally, our findings confirm that reported trust and reliance behavior are distinct constructs that should be evaluated separately in AI-assisted decision making. --- ### The Geometric Inductive Bias of Grokking: Bypassing Phase Transitions via Architectural Topology Authors: Alper Yıldırım Published: 2026-03-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.05228v3 Mechanistic interpretability typically relies on post-hoc analysis of trained networks. We instead adopt an interventional approach: testing hypotheses a priori by modifying architectural topology to observe training dynamics. We study grokking - delayed generalization in Transformers trained on cyclic modular addition (Zp) - investigating if specific architectural degrees of freedom prolong the memorization phase. We identify two independent structural factors in standard Transformers: unbounded representational magnitude and data-dependent attention routing. First, we introduce a fully bounded spherical topology enforcing L2 normalization throughout the residual stream and an unembedding matrix with a fixed temperature scale. This removes magnitude-based degrees of freedom, reducing grokking onset time by over 20x without weight decay. Second, a Uniform Attention Ablation overrides data-dependent query-key routing with a uniform distribution, reducing the attention layer to a Continuous Bag-of-Words (CBOW) aggregator. Despite removing adaptive routing, these models achieve 100% generalization across all seeds and bypass the grokking delay entirely. To evaluate whether this acceleration is a task-specific geometric alignment rather than a generic optimization stabilizer, we use non-commutative S5 permutation composition as a negative control. Enforcing spherical constraints on S5 does not accelerate generalization. This suggests eliminating the memorization phase depends strongly on aligning architectural priors with the task's intrinsic symmetries. Together, these findings provide interventional evidence that architectural degrees of freedom substantially influence grokking, suggesting a predictive structural perspective on training dynamics. --- ### Learning Optimal Individualized Decision Rules with Conditional Demographic Parity Authors: Wenhai Cui, Wen Su, Donglin Zeng, Xingqiu Zhao Published: 2026-03-05 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2603.05226v1 Individualized decision rules (IDRs) have become increasingly prevalent in societal applications such as personalized marketing, healthcare, and public policy design. However, a critical ethical concern arises from the potential discriminatory effects of IDRs trained on biased data. These algorithms may disproportionately harm individuals from minority subgroups defined by sensitive attributes like gender, race, or language. To address this issue, we propose a novel framework that incorporates demographic parity (DP) and conditional demographic parity (CDP) constraints into the estimation of optimal IDRs. We show that the theoretically optimal IDRs under DP and CDP constraints can be obtained by applying perturbations to the unconstrained optimal IDRs, enabling a computationally efficient solution. Theoretically, we derive convergence rates for both policy value and the fairness constraint term. The effectiveness of our methods is illustrated through comprehensive simulation studies and an empirical application to the Oregon Health Insurance Experiment. --- ### AI+HW 2035: Shaping the Next Decade Authors: Deming Chen, Jason Cong, Azalia Mirhoseini, Christos Kozyrakis, Subhasish Mitra, Jinjun Xiong, Cliff Young, Anima Anandkumar, Michael Littman, Aron Kirschen, Sophia Shao, Serge Leef et al. Published: 2026-03-05 Categories: cs.AI, cs.AR arXiv: https://arxiv.org/abs/2603.05225v1 Artificial intelligence (AI) and hardware (HW) are advancing at unprecedented rates, yet their trajectories have become inseparably intertwined. The global research community lacks a cohesive, long-term vision to strategically coordinate the development of AI and HW. This fragmentation constrains progress toward holistic, sustainable, and adaptive AI systems capable of learning, reasoning, and operating efficiently across cloud, edge, and physical environments. The future of AI depends not only on scaling intelligence, but on scaling efficiency, achieving exponential gains in intelligence per joule, rather than unbounded compute consumption. Addressing this grand challenge requires rethinking the entire computing stack. This vision paper lays out a 10-year roadmap for AI+HW co-design and co-development, spanning algorithms, architectures, systems, and sustainability. We articulate key insights that redefine scaling around energy efficiency, system-level integration, and cross-layer optimization. We identify key challenges and opportunities, candidly assess potential obstacles and pitfalls, and propose integrated solutions grounded in algorithmic innovation, hardware advances, and software abstraction. Looking ahead, we define what success means in 10 years: achieving a 1000x improvement in efficiency for AI training and inference; enabling energy-aware, self-optimizing systems that seamlessly span cloud, edge, and physical AI; democratizing access to advanced AI infrastructure; and embedding human-centric principles into the design of intelligent systems. Finally, we outline concrete action items for academia, industry, government, and the broader community, calling for coordinated national initiatives, shared infrastructure, workforce development, cross-agency collaboration, and sustained public-private partnerships to ensure that AI+HW co-design becomes a unifying long-term mission. --- ### SPyCer: Semi-Supervised Physics-Guided Contextual Attention for Near-Surface Air Temperature Estimation from Satellite Imagery Authors: Sofiane Bouaziz, Adel Hafiane, Raphael Canals, Rachid Nedjai Published: 2026-03-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.05219v1 Modern Earth observation relies on satellites to capture detailed surface properties. Yet, many phenomena that affect humans and ecosystems unfold in the atmosphere close to the surface. Near-ground sensors provide accurate measurements of certain environmental characteristics, such as near-surface air temperature (NSAT). However, they remain sparse and unevenly distributed, limiting their ability to provide continuous spatial measurements. To bridge this gap, we introduce SPyCer, a semi-supervised physics-guided network that can leverage pixel information and physical modeling to guide the learning process through meaningful physical properties. It is designed for continuous estimation of NSAT by proxy using satellite imagery. SPyCer frames NSAT prediction as a pixel-wise vision problem, where each near-ground sensor is projected onto satellite image coordinates and positioned at the center of a local image patch. The corresponding sensor pixel is supervised using both observed NSAT and physics-based constraints, while surrounding pixels contribute through physics-guided regularization derived from the surface energy balance and advection-diffusion-reaction partial differential equations. To capture the physical influence of neighboring pixels, SPyCer employs a multi-head attention guided by land cover characteristics and modulated with Gaussian distance weighting. Experiments on real-world datasets demonstrate that SPyCer produces spatially coherent and physically consistent NSAT estimates, outperforming existing baselines in terms of accuracy, generalization, and alignment with underlying physical processes. --- ### KARL: Knowledge Agents via Reinforcement Learning Authors: Jonathan D. Chang, Andrew Drozdov, Shubham Toshniwal, Owen Oertell, Alexander Trott, Jacob Portes, Abhay Gupta, Pallavi Koppol, Ashutosh Baheti, Sean Kulinski, Ivan Zhou, Irene Dea et al. Published: 2026-03-05 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.05218v1 We present a system for training enterprise search agents via reinforcement learning that achieves state-of-the-art performance across a diverse suite of hard-to-verify agentic search tasks. Our work makes four core contributions. First, we introduce KARLBench, a multi-capability evaluation suite spanning six distinct search regimes, including constraint-driven entity search, cross-document report synthesis, tabular numerical reasoning, exhaustive entity retrieval, procedural reasoning over technical documentation, and fact aggregation over internal enterprise notes. Second, we show that models trained across heterogeneous search behaviors generalize substantially better than those optimized for any single benchmark. Third, we develop an agentic synthesis pipeline that employs long-horizon reasoning and tool use to generate diverse, grounded, and high-quality training data, with iterative bootstrapping from increasingly capable models. Fourth, we propose a new post-training paradigm based on iterative large-batch off-policy RL that is sample efficient, robust to train-inference engine discrepancies, and naturally extends to multi-task training with out-of-distribution generalization. Compared to Claude 4.6 and GPT 5.2, KARL is Pareto-optimal on KARLBench across cost-quality and latency-quality trade-offs, including tasks that were out-of-distribution during training. With sufficient test-time compute, it surpasses the strongest closed models. These results show that tailored synthetic data in combination with multi-task reinforcement learning enables cost-efficient and high-performing knowledge agents for grounded reasoning. --- ### Early Warning of Intraoperative Adverse Events via Transformer-Driven Multi-Label Learning Authors: Xueyao Wang, Xiuding Cai, Honglin Shang, Yaoyao Zhu, Yu Yao Published: 2026-03-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.05212v2 Early warning of intraoperative adverse events plays a vital role in reducing surgical risk and improving patient safety. While deep learning has shown promise in predicting the single adverse event, several key challenges remain: overlooking adverse event dependencies, underutilizing heterogeneous clinical data, and suffering from the class imbalance inherent in medical datasets. To address these issues, we construct the first Multi-label Adverse Events dataset (MuAE) for intraoperative adverse events prediction, covering six critical events. Next, we propose a novel Transformerbased multi-label learning framework (IAENet) that combines an improved Time-Aware Feature-wise Linear Modulation (TAFiLM) module for static covariates and dynamic variables robust fusion and complex temporal dependencies modeling. Furthermore, we introduce a Label-Constrained Reweighting Loss (LCRLoss) with co-occurrence regularization to effectively mitigate intra-event imbalance and enforce structured consistency among frequently co-occurring events. Extensive experiments demonstrate that IAENet consistently outperforms strong baselines on 5, 10, and 15-minute early warning tasks, achieving improvements of +5.05%, +2.82%, and +7.57% on average F1 score. These results highlight the potential of IAENet for supporting intelligent intraoperative decision-making in clinical practice. --- ### Balancing Coverage and Draft Latency in Vocabulary Trimming for Faster Speculative Decoding Authors: Ofir Ben Shoham Published: 2026-03-05 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.05210v1 Speculative decoding accelerates inference for Large Language Models by using a lightweight draft model to propose candidate tokens that are verified in parallel by a larger target model. Prior work shows that the draft model often dominates speculative decoding latency, since it generates tokens sequentially and incurs high cost from its language modeling head as vocabulary size grows. This exposes a fundamental trade-off in draft model design: larger vocabularies improve token coverage and agreement with the target model, but incur higher draft latency, while smaller vocabularies reduce latency at the risk of missing tokens required for accurate draft generation. We address this trade-off through vocabulary trimming for draft models, motivated by the observation that domain-specific workloads use only a small fraction of the full vocabulary. We cast draft vocabulary selection as a constrained optimization problem that balances token coverage and draft latency. Coverage is computed over assistant responses in the training data, while latency is estimated using architecture-aware FLOPs that capture the cost of the language modeling head as a function of vocabulary size. We optimize a utility function with a Tree-structured Parzen Estimator to efficiently explore the coverage-latency Pareto frontier under a minimum coverage constraint. Experiments show improved speculative decoding throughput while reducing draft vocabularies by up to 97% with high coverage. On domain-specific tasks, we achieve up to 16% latency reduction and 20% throughput improvement, and up to 6.7% throughput gains on diverse out-of-distribution tasks. --- ### Core-based Hierarchies for Efficient GraphRAG Authors: Jakir Hossain, Ahmet Erdem Sarıyüce Published: 2026-03-05 Categories: cs.IR, cs.CL arXiv: https://arxiv.org/abs/2603.05207v2 Retrieval-Augmented Generation (RAG) enhances large language models by incorporating external knowledge. However, existing vector-based methods often fail on global sensemaking tasks that require reasoning across many documents. GraphRAG addresses this by organizing documents into a knowledge graph with hierarchical communities that can be recursively summarized. Current GraphRAG approaches rely on Leiden clustering for community detection, but we prove that on sparse knowledge graphs, where average degree is constant and most nodes have low degree, modularity optimization admits exponentially many near-optimal partitions, making Leiden-based communities inherently non-reproducible. To address this, we propose replacing Leiden with k-core decomposition, which yields a deterministic, density-aware hierarchy in linear time. We introduce a set of lightweight heuristics that leverage the k-core hierarchy to construct size-bounded, connectivity-preserving communities for retrieval and summarization, along with a token-budget-aware sampling strategy that reduces LLM costs. We evaluate our methods on real-world datasets including financial earnings transcripts, news articles, and podcasts, using three LLMs for answer generation and five independent LLM judges for head-to-head evaluation. Across datasets and models, our approach consistently improves answer comprehensiveness and diversity while reducing token usage, demonstrating that k-core-based GraphRAG is an effective and efficient framework for global sensemaking. --- ### Stable-LoRA: Stabilizing Feature Learning of Low-Rank Adaptation Authors: Yize Wu, Ke Gao, Ling Li, Yanjun Wu Published: 2026-03-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.05204v1 Low-Rank Adaptation (LoRA) is a widely adopted parameter-efficient method for fine-tuning Large Langauge Models. It updates the weight matrix as $W=W_0+sBA$, where $W_0$ is the original frozen weight, $s$ is a scaling factor and $A$,$B$ are trainable low-rank matrices. Despite its robust empirical effectiveness, the theoretical foundations of LoRA remain insufficiently understood, particularly with respect to feature learning stability. In this paper, we first establish that, LoRA can, in principle, naturally achieve and sustain stable feature learning (i.e., be self-stabilized) under appropriate hyper-parameters and initializations of $A$ and $B$. However, we also uncover a fundamental limitation that the necessary non-zero initialization of $A$ compromises self-stability, leading to suboptimal performances. To address this challenge, we propose Stable-LoRA, a weight-shrinkage optimization strategy that dynamically enhances stability of LoRA feature learning. By progressively shrinking $A$ during the earliest training steps, Stable-LoRA is both theoretically and empirically validated to effectively eliminate instability of LoRA feature learning while preserving the benefits of the non-zero start. Experiments show that Stable-LoRA consistently outperforms other baselines across diverse models and tasks, with no additional memory usage and only negligible computation overheads. The code is available at https://github.com/Yize-Wu/Stable-LoRA. --- ### Semantic Class Distribution Learning for Debiasing Semi-Supervised Medical Image Segmentation Authors: Yingxue Su, Yiheng Zhong, Keying Zhu, Zimu Zhang, Zhuoru Zhang, Yifang Wang, Yuxin Zhang, Xinyuan Zheng, Jingxin Liu, Xiaofeng Liu Published: 2026-03-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.05202v2 Medical image segmentation is critical for computer-aided diagnosis. However, dense pixel-level annotation is time-consuming and costly, and medical datasets often exhibit severe class imbalance. Such an imbalance causes minority structures to be overwhelmed by dominant classes in feature representations, hindering the learning of discriminative features and making reliable segmentation particularly challenging. To address this, we propose the Semantic Class Distribution Learning (SCDL) framework, a plug-and-play module that mitigates supervision and representation biases by learning structured class-conditional feature distributions. SCDL integrates Class Distribution Bidirectional Alignment (CDBA) to align embeddings with learnable class proxies and leverages Semantic Anchor Constraints (SAC) to guide proxies using labeled data. Experiments on the Synapse and AMOS datasets demonstrate that SCDL largely improves segmentation performance across both overall and class-level metrics, with particularly notable gains for several low-frequency organs. Our anonymous code is released at https://anonymous.4open.science/r/SCDL. --- ### Semantic Class Distribution Learning for Debiasing Semi-Supervised Medical Image Segmentation Authors: Yingxue Su, Yiheng Zhong, Keying Zhu, Zimu Zhang, Zhuoru Zhang, Yifang Wang, Yuxin Zhang, Jingxin Liu Published: 2026-03-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.05202v1 Medical image segmentation is critical for computer-aided diagnosis. However, dense pixel-level annotation is time-consuming and expensive, and medical datasets often exhibit severe class imbalance. Such imbalance causes minority structures to be overwhelmed by dominant classes in feature representations, hindering the learning of discriminative features and making reliable segmentation particularly challenging. To address this, we propose the Semantic Class Distribution Learning (SCDL) framework, a plug-and-play module that mitigates supervision and representation biases by learning structured class-conditional feature distributions. SCDL integrates Class Distribution Bidirectional Alignment (CDBA) to align embeddings with learnable class proxies and leverages Semantic Anchor Constraints (SAC) to guide proxies using labeled data. Experiments on the Synapse and AMOS datasets demonstrate that SCDL significantly improves segmentation performance across both overall and class-level metrics, with particularly strong gains on minority classes, achieving state-of-the-art results. Our code is released at https://github.com/Zyh55555/SCDL. --- ### Towards a data-scale independent regulariser for robust sparse identification of non-linear dynamics Authors: Jay Raut, Daniel N. Wilke, Stephan Schmidt Published: 2026-03-05 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2603.05201v1 Data normalisation, a common and often necessary preprocessing step in engineering and scientific applications, can severely distort the discovery of governing equations by magnitudebased sparse regression methods. This issue is particularly acute for the Sparse Identification of Nonlinear Dynamics (SINDy) framework, where the core assumption of sparsity is undermined by the interaction between data scaling and measurement noise. The resulting discovered models can be dense, uninterpretable, and physically incorrect. To address this critical vulnerability, we introduce the Sequential Thresholding of Coefficient of Variation (STCV), a novel, computationally efficient sparse regression algorithm that is inherently robust to data scaling. STCV replaces conventional magnitude-based thresholding with a dimensionless statistical metric, the Coefficient Presence (CP), which assesses the statistical validity and consistency of candidate terms in the model library. This shift from magnitude to statistical significance makes the discovery process invariant to arbitrary data scaling. Through comprehensive benchmarking on canonical dynamical systems and practical engineering problems, including a physical mass-spring-damper experiment, we demonstrate that STCV consistently and significantly outperforms standard Sequential Thresholding Least Squares (STLSQ) and Ensemble-SINDy (E-SINDy) on normalised, noisy datasets. The results show that STCV-based methods can successfully identify the correct, sparse physical laws even when other methods fail. By mitigating the distorting effects of normalisation, STCV makes sparse system identification a more reliable and automated tool for real-world applications, thereby enhancing model interpretability and trustworthiness. --- ### Distilling Formal Logic into Neural Spaces: A Kernel Alignment Approach for Signal Temporal Logic Authors: Sara Candussio, Gabriele Sarti, Gaia Saveri, Luca Bortolussi Published: 2026-03-05 Categories: cs.CL, cs.SC arXiv: https://arxiv.org/abs/2603.05198v1 We introduce a framework for learning continuous neural representations of formal specifications by distilling the geometry of their semantics into a latent space. Existing approaches rely either on symbolic kernels -- which preserve behavioural semantics but are computationally prohibitive, anchor-dependent, and non-invertible -- or on syntax-based neural embeddings that fail to capture underlying structures. Our method bridges this gap: using a teacher-student setup, we distill a symbolic robustness kernel into a Transformer encoder. Unlike standard contrastive methods, we supervise the model with a continuous, kernel-weighted geometric alignment objective that penalizes errors in proportion to their semantic discrepancies. Once trained, the encoder produces embeddings in a single forward pass, effectively mimicking the kernel's logic at a fraction of its computational cost. We apply our framework to Signal Temporal Logic (STL), demonstrating that the resulting neural representations faithfully preserve the semantic similarity of STL formulae, accurately predict robustness and constraint satisfaction, and remain intrinsically invertible. Our proposed approach enables highly efficient, scalable neuro-symbolic reasoning and formula reconstruction without repeated kernel computation at runtime. --- ### Masked diffusion LLMs can use EoS tokens for hidden reasoning Authors: Sarah Breckner, Sebastian Schuster Published: 2026-03-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.05197v2 Diffusion LLMs have been proposed as an alternative to autoregressive LLMs. Curiously, they are especially capable if the generation length, i.e., the number of tokens the model has to output, is set to a much higher value than the correct answer length, and the model pads its answer with end-of-sequence (EoS) tokens. We hypothesize that off-the-shelf masked diffusion LLMs use the representations of EoS tokens as additional computing capacity, which enhances their performance. We experiment with the diffusion models LLaDA1.5, LLaDA2.0-mini, and Dream-v0 on three reasoning tasks: Addition, Entity Tracking, and Sudoku. In a controlled prompting experiment, we confirm that adding EoS tokens improves the LLMs' performance. To further verify whether their representations are used for hidden computations, we perform a causal intervention and transfer the hidden states of the EoS tokens between generations, which increases the models' relative likelihood of outputting the counterfactual answer. The behavioral experiments and the causal interventions indicate that fully bidirectional masked diffusion LLMs can indeed perform latent reasoning in the representations of EoS tokens. Furthermore, we find that these results generalize beyond toy tasks and that providing the model with additional EoS tokens also improves performance on GSM8K and two-hop reasoning. --- ### Diffusion LLMs can think EoS-by-EoS Authors: Sarah Breckner, Sebastian Schuster Published: 2026-03-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.05197v1 Diffusion LLMs have been proposed as an alternative to autoregressive LLMs, excelling especially at complex reasoning tasks with interdependent sub-goals. Curiously, this is particularly true if the generation length, i.e., the number of tokens the model has to output, is set to a much higher value than is required for providing the correct answer to the task, and the model pads its answer with end-of-sequence (EoS) tokens. We hypothesize that diffusion models think EoS-by-EoS, that is, they use the representations of EoS tokens as a hidden scratchpad, which allows them to solve harder reasoning problems. We experiment with the diffusion models LLaDA1.5, LLaDA2.0-mini, and Dream-v0 on the tasks Addition, Entity Tracking, and Sudoku. In a controlled prompting experiment, we confirm that adding EoS tokens improves the LLMs' reasoning capabilities. To further verify whether they serve as space for hidden computations, we patch the hidden states of the EoS tokens with those of a counterfactual generation, which frequently changes the generated output to the counterfactual. The success of the causal intervention underscores that the EoS tokens, which one may expect to be devoid of meaning, carry information on the problem to solve. The behavioral experiments and the causal interventions indicate that diffusion LLMs can indeed think EoS-by-EoS. --- ### Transducing Language Models Authors: Vésteinn Snæbjarnarson, Samuel Kiegeland, Tianyu Liu, Reda Boumasmoud, Ryan Cotterell, Tim Vieira Published: 2026-03-05 Categories: cs.CL, cs.FL arXiv: https://arxiv.org/abs/2603.05193v1 Modern language models define distributions over strings, but downstream tasks often require different output formats. For instance, a model that generates byte-pair strings does not directly produce word-level predictions, and a DNA model does not directly produce amino-acid sequences. In such cases, a deterministic string-to-string transformation can convert the model's output to the desired form. This is a familiar pattern in probability theory: applying a function $f$ to a random variable $X\sim p$ yields a transformed random variable $f(X)$ with an induced distribution. While such transformations are occasionally used in language modeling, prior work does not treat them as yielding new, fully functional language models. We formalize this perspective and introduce a general framework for language models derived from deterministic string-to-string transformations. We focus on transformations representable as finite-state transducers -- a commonly used state-machine abstraction for efficient string-to-string mappings. We develop algorithms that compose a language model with an FST to *marginalize* over source strings mapping to a given target, propagating probabilities through the transducer without altering model parameters and enabling *conditioning* on transformed outputs. We present an exact algorithm, an efficient approximation, and a theoretical analysis. We conduct experiments in three domains: converting language models from tokens to bytes, from tokens to words, and from DNA to amino acids. These experiments demonstrate inference-time adaptation of pretrained language models to match application-specific output requirements. --- ### Escaping the Hydrolysis Trap: An Agentic Workflow for Inverse Design of Durable Photocatalytic Covalent Organic Frameworks Authors: Iman Peivaste, Nicolas D. Boscher, Ahmed Makradi, Salim Belouettar Published: 2026-03-05 Categories: physics.chem-ph, cond-mat.mtrl-sci, cs.AI, physics.comp-ph arXiv: https://arxiv.org/abs/2603.05188v1 Covalent organic frameworks (COFs) are promising photocatalysts for solar hydrogen production, yet the most electronically favorable linkages, imines, hydrolyze rapidly in water, creating a stability--activity trade-off that limits practical deployment. Navigating the combinatorial design space of nodes, linkers, linkages, and functional groups to identify candidates that are simultaneously active and durable remains a formidable challenge. Here we introduce Ara, a large-language-model (LLM) agent that leverages pretrained chemical knowledge, donor--acceptor theory, conjugation effects, and linkage stability hierarchies, to guide the search for photocatalytic COFs satisfying joint band-gap, band-edge, and hydrolytic-stability criteria. Evaluated against random search and Bayesian optimization (BO) over a space consisting of candidates with various nodes, linkers, linkages, and r-groups, screened with a GFN1-xTB fragment pipeline, Ara achieves a 52.7\% hit rate (11.5$\times$ random, p = 0.006), finds its first hit at iteration 12 versus 25 for random search, and significantly outperforms BO (p = 0.006). Inspection of the agent's reasoning traces reveals interpretable chemical logic: early convergence on vinylene and beta-ketoenamine linkages for stability, node selection informed by electron-withdrawing character, and systematic R-group optimization to center the band gap at 2.0 eV. Exhaustive evaluation of the full search space uncovers a complementary exploitation--exploration trade-off between the agent and BO, suggesting that hybrid strategies may combine the strengths of both approaches. These results demonstrate that LLM chemical priors can substantially accelerate multi-criteria materials discovery. --- ### Logi-PAR: Logic-Infused Patient Activity Recognition via Differentiable Rule Authors: Muhammad Zarar, MingZheng Zhang, Xiaowang Zhang, Zhiyong Feng, Sofonias Yitagesu, Kawsar Farooq Published: 2026-03-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.05184v1 Patient Activity Recognition (PAR) in clinical settings uses activity data to improve safety and quality of care. Although significant progress has been made, current models mainly identify which activity is occurring. They often spatially compose sub-sparse visual cues using global and local attention mechanisms, yet only learn logically implicit patterns due to their neural-pipeline. Advancing clinical safety requires methods that can infer why a set of visual cues implies a risk, and how these can be compositionally reasoned through explicit logic beyond mere classification. To address this, we proposed Logi-PAR, the first Logic-Infused Patient Activity Recognition Framework that integrates contextual fact fusion as a multi-view primitive extractor and injects neural-guided differentiable rules. Our method automatically learns rules from visual cues, optimizing them end-to-end while enabling the implicit emergence patterns to be explicitly labelled during training. To the best of our knowledge, Logi-PAR is the first framework to recognize patient activity by applying learnable logic rules to symbolic mappings. It produces auditable why explanations as rule traces and supports counterfactual interventions (e.g., risk would decrease by 65% if assistance were present). Extensive evaluation on clinical benchmarks (VAST and OmniFall) demonstrates state-of-the-art performance, significantly outperforming Vision-Language Models and transformer baselines. The code is available via: https://github.com/zararkhan985/Logi-PAR.git} --- ### Mario: Multimodal Graph Reasoning with Large Language Models Authors: Yuanfu Sun, Kang Li, Pengkang Guo, Jiajin Liu, Qiaoyu Tan Published: 2026-03-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.05181v2 Recent advances in large language models (LLMs) have opened new avenues for multimodal reasoning. Yet, most existing methods still rely on pretrained vision-language models (VLMs) to encode image-text pairs in isolation, ignoring the relational structure that real-world multimodal data naturally form. This motivates reasoning on multimodal graphs (MMGs), where each node has textual and visual attributes and edges provide structural cues. Enabling LLM-based reasoning on such heterogeneous multimodal signals while preserving graph topology introduces two key challenges: resolving weak cross-modal consistency and handling heterogeneous modality preference. To address this, we propose Mario, a unified framework that simultaneously resolves the two above challenges and enables effective LLM-based reasoning over MMGs. Mario consists of two innovative stages. Firstly, a graph-conditioned VLM design that jointly refines textual and visual features through fine-grained cross-modal contrastive learning guided by graph topology. Secondly, a modality-adaptive graph instruction tuning mechanism that organizes aligned multimodal features into graph-aware instruction views and employs a learnable router to surface, for each node and its neighborhood, the most informative modality configuration to the LLM. Extensive experiments across diverse MMG benchmarks demonstrate that Mario consistently outperforms state-of-the-art graph models in both supervised and zero-shot scenarios for node classification and link prediction. The code will be made available at https://github.com/sunyuanfu/Mario. --- ### Incentive Aware AI Regulations: A Credal Characterisation Authors: Anurag Singh, Julian Rodemann, Rajeev Verma, Siu Lun Chau, Krikamol Muandet Published: 2026-03-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.05175v2 The rapid proliferation of AI applications has intensified debate on effective regulation of these black-box services. Effective regulation must balance two competing goals: (1) deterring non-compliant providers from entering the market, while (2) retaining compliant ones. We call this ideal the perfect market outcome (PMO). Regulators face two compounding obstacles that make PMO difficult to achieve: providers hold private information and can act strategically to evade compliance, while any evidence drawn or derived from a finite sample carries statistical uncertainty in proving non-compliance. As this information asymmetry and statistical uncertainty is inherent to any effective regulation, we formalise them through a mechanism design framework that explicitly accounts for such statistical uncertainty. This yields a sharp characterisation: a mechanism achieves PMO if and only if the set of non-compliant evidence distributions forms a closed, convex set of probability measures, known in imprecise probability as a credal set. This result serves as a diagnostic tool to determine whether PMO is achievable under a given regulation. We further show that PMO-achieving mechanisms can be constructed from a collection of hypothesis tests, and validate our theoretical contributions through experiments on spurious-feature and fairness-based regulations. --- ### Trainable Bitwise Soft Quantization for Input Feature Compression Authors: Karsten Schrödter, Jan Stenkamp, Nina Herrmann, Fabian Gieseke Published: 2026-03-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.05172v1 The growing demand for machine learning applications in the context of the Internet of Things calls for new approaches to optimize the use of limited compute and memory resources. Despite significant progress that has been made w.r.t. reducing model sizes and improving efficiency, many applications still require remote servers to provide the required resources. However, such approaches rely on transmitting data from edge devices to remote servers, which may not always be feasible due to bandwidth, latency, or energy constraints. We propose a task-specific, trainable feature quantization layer that compresses the input features of a neural network. This can significantly reduce the amount of data that needs to be transferred from the device to a remote server. In particular, the layer allows each input feature to be quantized to a user-defined number of bits, enabling a simple on-device compression at the time of data collection. The layer is designed to approximate step functions with sigmoids, enabling trainable quantization thresholds. By concatenating outputs from multiple sigmoids, introduced as bitwise soft quantization, it achieves trainable quantized values when integrated with a neural network. We compare our method to full-precision inference as well as to several quantization baselines. Experiments show that our approach outperforms standard quantization methods, while maintaining accuracy levels close to those of full-precision models. In particular, depending on the dataset, compression factors of $5\times$ to $16\times$ can be achieved compared to $32$-bit input without significant performance loss. --- ### Guidelines for the Annotation and Visualization of Legal Argumentation Structures in Chinese Judicial Decisions Authors: Kun Chen, Xianglei Liao, Kaixue Fei, Yi Xing, Xinrui Li Published: 2026-03-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.05171v2 This Guideline presents a systematic and operationalizable annotation framework for representing legal argumentation structures in judicial decisions. Grounded in theories of legal reasoning and argumentation, the framework aims to reveal the logical organization of judicial reasoning and provide a reliable foundation for computational analysis. At the element level, the Guideline distinguishes between the non-propositional layer and the propositional layer. The non-propositional layer consists of two elements: Issue and Non-argumentative Component. At the propositional level, the Guideline defines four proposition types: General Normative Judgment, Particular Normative Judgment, General Factual Judgment, and Particular Factual Judgment. At the relational level, five relation types are defined to represent argumentative structures: Support, Attack, Joint, Match, and Identity. These relations capture positive and negative argumentative connections, conjunctive reasoning structures, correspondences between legal norms and case facts, and identity or semantic equivalence between propositions. The Guideline further specifies formal representation rules and visualization conventions for both basic and nested structures, enabling consistent visualization of complex argumentation patterns. In addition, it establishes a standardized annotation workflow and consistency control mechanisms to ensure the reproducibility and reliability of annotated data. By providing a clear conceptual model, formal representation rules, and practical annotation procedures, this Guideline supports large-scale analysis of judicial reasoning and future research in legal argument mining, computational modeling of legal reasoning, and AI-assisted legal analysis. --- ### Sparse-BitNet: 1.58-bit LLMs are Naturally Friendly to Semi-Structured Sparsity Authors: Di Zhang, Xun Wu, Shaohan Huang, Yudong Wang, Hanyong Shao, Yingbo Hao, Zewen Chi, Li Dong, Ting Song, Yan Xia, Zhifang Sui, Furu Wei Published: 2026-03-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.05168v1 Semi-structured N:M sparsity and low-bit quantization (e.g., 1.58-bit BitNet) are two promising approaches for improving the efficiency of large language models (LLMs), yet they have largely been studied in isolation. In this work, we investigate their interaction and show that 1.58-bit BitNet is naturally more compatible with N:M sparsity than full-precision models. To study this effect, we propose Sparse-BitNet, a unified framework that jointly applies 1.58-bit quantization and dynamic N:M sparsification while ensuring stable training for the first time. Across multiple model scales and training regimes (sparse pretraining and dense-to-sparse schedules), 1.58-bit BitNet consistently exhibits smaller performance degradation than full-precision baselines at the same sparsity levels and can tolerate higher structured sparsity before accuracy collapse. Moreover, using our custom sparse tensor core, Sparse-BitNet achieves substantial speedups in both training and inference, reaching up to 1.30X. These results highlight that combining extremely low-bit quantization with semi-structured N:M sparsity is a promising direction for efficient LLMs. Code available at https://github.com/AAzdi/Sparse-BitNet --- ### C2-Faith: Benchmarking LLM Judges for Causal and Coverage Faithfulness in Chain-of-Thought Reasoning Authors: Avni Mittal, Rauno Arike Published: 2026-03-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.05167v2 Large language models (LLMs) are increasingly used as judges of chain-of-thought (CoT) reasoning, yet it remains unclear whether they can reliably assess process faithfulness rather than merely answer plausibility. We introduce C2-Faith, a benchmark built from PRM800K that explicitly decomposes faithfulness into two complementary dimensions: causality (whether each step logically follows from prior context) and coverage (whether essential intermediate inferences are present). Using controlled perturbations, we construct examples with known causal error positions by replacing a single step with a logically inconsistent variant, and with controlled coverage deletions at varying rates, enabling direct measurement against reference labels. We evaluate three frontier LLM judges across three tasks: binary causal detection, causal step localization, and coverage scoring. Our results reveal that judge reliability is highly task-dependent, with no single model dominating across settings. While models often detect that an error exists, they struggle to accurately localize it, indicating a substantial gap between detection and attribution. Moreover, all judges systematically overestimate reasoning completeness, assigning high coverage scores even when substantial portions of intermediate reasoning are missing. These findings expose fundamental limitations of LLM judges in process-level evaluation and highlight the need for more reliable and calibrated methods when using LLMs to assess reasoning quality. --- ### A Geometry-Adaptive Deep Variational Framework for Phase Discovery in the Landau-Brazovskii Model Authors: Yuchen Xie, Jianyuan Yin, Lei Zhang Published: 2026-03-05 Categories: cond-mat.mtrl-sci, cs.LG arXiv: https://arxiv.org/abs/2603.05161v1 The discovery of ordered structures in pattern-forming systems, such as the Landau-Brazovskii (LB) model, is often limited by the sensitivity of numerical solvers to the prescribed computational domain size. Incompatible domains induce artificial stress, frequently trapping the system in high-energy metastable configurations. To resolve this issue, we propose a Geometry-Adaptive Deep Variational Framework (GeoDVF) that jointly optimizes the infinite-dimensional order parameter, which is parameterized by a neural network, and the finite-dimensional geometric parameters of the computational domain. By explicitly treating the domain size as trainable variables within the variational formulation, GeoDVF naturally eliminates artificial stress during training. To escape the attraction basin of the disordered phase under small initializations, we introduce a warmup penalty mechanism, which effectively destabilizes the disordered phase, enabling the spontaneous nucleation of complex three-dimensional ordered phases from random initializations. Furthermore, we design a guided initialization protocol to resolve topologically intricate phases associated with narrow basins of attraction. Extensive numerical experiments show that GeoDVF provides a robust and geometry-consistent variational solver capable of identifying both stable and metastable states without prior knowledge. --- ### Lifelong Language-Conditioned Robotic Manipulation Learning Authors: Xudong Wang, Zebin Han, Zhiyu Liu, Gan Li, Jiahua Dong, Baichen Liu, Lianqing Liu, Zhi Han Published: 2026-03-05 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2603.05160v1 Traditional language-conditioned manipulation agent sequential adaptation to new manipulation skills leads to catastrophic forgetting of old skills, limiting dynamic scene practical deployment. In this paper, we propose SkillsCrafter, a novel robotic manipulation framework designed to continually learn multiple skills while reducing catastrophic forgetting of old skills. Specifically, we propose a Manipulation Skills Adaptation to retain the old skills knowledge while inheriting the shared knowledge between new and old skills to facilitate learning of new skills. Meanwhile, we perform the singular value decomposition on the diverse skill instructions to obtain common skill semantic subspace projection matrices, thereby recording the essential semantic space of skills. To achieve forget-less and generalization manipulation, we propose a Skills Specialization Aggregation to compute inter-skills similarity in skill semantic subspaces, achieving aggregation of the previously learned skill knowledge for any new or unknown skill. Extensive experiments demonstrate the effectiveness and superiority of our proposed SkillsCrafter. ---