# Research Papers — 2026-03 · bundle page 131 of 243 > Full-text markdown of papers #6501–6550 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/131 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/130.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/132.md) ### PARSA-Bench: A Comprehensive Persian Audio-Language Model Benchmark Authors: Mohammad Javad Ranjbar Kalahroodi, Mohammad Amini, Parmis Bathayan, Heshaam Faili, Azadeh Shakery Published: 2026-03-15 Categories: cs.CL, cs.SD arXiv: https://arxiv.org/abs/2603.14456v2 Persian poses unique audio understanding challenges through its classical poetry, traditional music, and pervasive code-switching, none of which is captured by existing benchmarks. We introduce \textbf{PARSA-Bench} (\textbf{P}ersian \textbf{A}udio \textbf{R}easoning and \textbf{S}peech \textbf{A}ssessment Benchmark), the first dedicated benchmark for evaluating LALMs on Persian language and culture. It covers 16 tasks, ten of them new, spanning speech understanding, paralinguistic analysis, and culturally grounded audio reasoning. Across most tasks, text-only baselines outperform their audio counterparts, so audio understanding rather than language knowledge remains the main limitation, and supplying the transcript alongside the audio lifts weak models to near their text-only level. The consistent exception is Persian poetry, where prosody carries information the written form cannot: audio beats text on both poetry tasks, and metre detection shows the first signs of being learnable only at the largest model scale. The dataset is publicly available at: https://huggingface.co/datasets/MohammadJRanjbar/PARSA-Bench --- ### PARSA-Bench: A Comprehensive Persian Audio-Language Model Benchmark Authors: Mohammad Javad Ranjbar Kalahroodi, Mohammad Amini, Parmis Bathayan, Heshaam Faili, Azadeh Shakery Published: 2026-03-15 Categories: cs.CL, cs.SD arXiv: https://arxiv.org/abs/2603.14456v1 Persian poses unique audio understanding challenges through its classical poetry, traditional music, and pervasive code-switching - none captured by existing benchmarks. We introduce PARSA-Bench (Persian Audio Reasoning and Speech Assessment Benchmark), the first benchmark for evaluating large audio-language models on Persian language and culture, comprising 16 tasks and over 8,000 samples across speech understanding, paralinguistic analysis, and cultural audio understanding. Ten tasks are newly introduced, including poetry meter and style detection, traditional Persian music understanding, and code-switching detection. Text-only baselines consistently outperform audio counterparts, suggesting models may not leverage audio-specific information beyond what transcription alone provides. Culturally-grounded tasks expose a qualitatively distinct failure mode: all models perform near random chance on vazn detection regardless of scale, suggesting prosodic perception remains beyond the reach of current models. The dataset is publicly available at https://huggingface.co/datasets/MohammadJRanjbar/PARSA-Bench --- ### Uni-MDTrack: Learning Decoupled Memory and Dynamic States for Parameter-Efficient Visual Tracking in All Modality Authors: Wenrui Cai, Zhenyi Lu, Yuzhe Li, Yongchao Feng, Jinqing Zhang, Qingjie Liu, Yunhong Wang Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14452v1 With the advent of Transformer-based one-stream trackers that possess strong capability in inter-frame relation modeling, recent research has increasingly focused on how to introduce spatio-temporal context. However, most existing methods rely on a limited number of historical frames, which not only leads to insufficient utilization of the context, but also inevitably increases the length of input and incurs prohibitive computational overhead. Methods that query an external memory bank, on the other hand, suffer from inadequate fusion between the retrieved spatio-temporal features and the backbone. Moreover, using discrete historical frames as context overlooks the rich dynamics of the target. To address the issues, we propose Uni-MDTrack, which consists of two core components: Memory-Aware Compression Prompt (MCP) module and Dynamic State Fusion (DSF) module. MCP effectively compresses rich memory features into memory-aware prompt tokens, which deeply interact with the input throughout the entire backbone, significantly enhancing the performance while maintaining a stable computational load. DSF complements the discrete memory by capturing the continuous dynamic, progressively introducing the updated dynamic state features from shallow to deep layers, while also preserving high efficiency. Uni-MDTrack also supports unified tracking across RGB, RGB-D/T/E, and RGB-Language modalities. Experiments show that in Uni-MDTrack, training only the MCP, DSF, and prediction head, keeping the proportion of trainable parameters around 30%, yields substantial performance gains, achieves state-of-the-art results on 10 datasets spanning five modalities. Furthermore, both MCP and DSF exhibit excellent generality, functioning as plug-and-play components that can boost the performance of various baseline trackers, while significantly outperforming existing parameter-efficient training approaches. --- ### How to find expressible and trainable parameterized quantum circuits? Authors: Peter Röseler, Dennis Willsch, Kristel Michielsen Published: 2026-03-15 Categories: quant-ph, cs.LG arXiv: https://arxiv.org/abs/2603.14451v1 Whether parameterized quantum circuits (PQCs) can be systematically constructed to be both trainable and expressive remains an open question. Highly expressive PQCs often exhibit barren plateaus, while several trainable alternatives admit efficient classical simulation. We address this question by deriving a finite-sample, dimension-independent concentration bound for estimating the variance of a PQC cost function, yielding explicit trainability guarantees. Across commonly used ansätze, we observe an anticorrelation between trainability and expressibility, consistent with theoretical insights. Building on this observation, we propose a property-based ansatz-search framework for identifying circuits that combine trainability and expressibility. We demonstrate its practical viability on a real quantum computer and apply it to variational quantum algorithms. We identify quantum neural network ansätze with improved effective dimension using over $6 \times$ fewer parameters, and for VQE on $\mathrm{H}_2$ we achieve UCCSD-like accuracy at substantially reduced circuit complexity. --- ### Zoom to Essence: Trainless GUI Grounding by Inferring upon Interface Elements Authors: Ziwei Liu, Tao Feng, Borui Kang, Yanbing Yang, Jun Luo Published: 2026-03-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.14448v1 Multimodal Large Language Model (MLLM)-based Graphical User Interface (GUI) agents develop rapidly, with visual grounding that maps natural language instructions to target UI elements serving as the core capability. Existing GUI agents typically fine-tune MLLM on massive datasets to handle challenges in understanding instructions and UI interfaces, which not only incurs high data annotation costs but also makes performance dependent on data quality and distribution. To avoid such cumbersome yet ineffective training, we notice that complex UI interfaces can be decomposed into basic visual elements directly understandable by common MLLMs. Consequently, we propose ZoomUI that leverages inference scaling to guide common MLLMs in progressively anchor instruction elements to increasingly detailed interface elements. Specifically, ZoomUI first optimizes the latent thinking to transform original instruction into element visual features description, and subsequently leverages internal attention to iteratively zoom in target element interface region. Evaluations on extensive benchmarks demonstrate that ZoomUI reaches or even surpasses SOTA baselines. --- ### Echoes Across Centuries: Phonetic Signatures of Persian Poets Authors: Kourosh Shahnazari, Seyed Moein Ayyoubzadeh, Mohammadali Keshtparvar Published: 2026-03-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.14443v2 This study examines phonetic texture in Persian poetry as a literary-historical phenomenon rather than a by-product of meter or a feature used only for classification. The analysis draws on a large corpus of 1,116,306 mesras from 31,988 poems written by 83 poets, restricted to five major classical meters to enable controlled comparison. Each line is converted into a grapheme-to-phoneme representation and analyzed using six phonetic metrics: hardness, sonority, sibilance, vowel ratio, phoneme entropy, and consonant-cluster ratio. Statistical models estimate poet-level differences while controlling for meter, poetic form, and line length. The results show that although meter and form explain a substantial portion of phonetic variation, they do not eliminate systematic differences between poets. Persian poetic sound therefore appears as conditioned variation within shared prosodic structures rather than as either purely individual style or simple metrical residue. A multidimensional stylistic map reveals several recurrent phonetic profiles, including high-sonority lyric styles, hardness-driven rhetorical or epic styles, sibilant mystical contours, and high-entropy complex textures. Historical analysis indicates that phonetic distributions shift across centuries, reflecting changes in genre prominence, literary institutions, and performance contexts rather than abrupt stylistic breaks. The study establishes a corpus-scale framework for phonetic analysis in Persian poetry and demonstrates how computational phonetics can contribute to literary-historical interpretation while remaining attentive to the formal structures that shape Persian verse. --- ### AR-Flow VAE: A Structured Autoregressive Flow Prior Variational Autoencoder for Unsupervised Blind Source Separation Authors: Yuan-Hao Wei, Fu-Hao Deng, Lin-Yong Cui, Yan-Jie Sun Published: 2026-03-15 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2603.14441v2 Blind source separation (BSS) seeks to recover latent source signals from observed mixtures. Variational autoencoders (VAEs) offer a natural perspective for this problem: the latent variables can be interpreted as source components, the encoder can be viewed as a demixing mapping from observations to sources, and the decoder can be regarded as a remixing process from inferred sources back to observations. In this work, we propose AR-Flow VAE, a novel VAE-based framework for BSS in which each latent source is endowed with a parameter-adaptive autoregressive flow prior. This prior significantly enhances the flexibility of latent source modeling, enabling the framework to capture complex non-Gaussian behaviors and structured dependencies, such as temporal correlations, that are difficult to represent with conventional priors. In addition, the structured prior design assigns distinct priors to different latent dimensions, thereby encouraging the latent components to separate into different source signals under heterogeneous prior constraints. Experimental results validate the effectiveness of the proposed architecture for blind source separation. More importantly, this work provides a foundation for future investigations into the identifiability and interpretability of AR-Flow VAE. --- ### End-to-End Spatial-Temporal Transformer for Real-time 4D HOI Reconstruction Authors: Haoyu Zhang, Wei Zhai, Yuhang Yang, Yang Cao, Zheng-Jun Zha Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14435v1 Monocular 4D human-object interaction (HOI) reconstruction - recovering a moving human and a manipulated object from a single RGB video - remains challenging due to depth ambiguity and frequent occlusions. Existing methods often rely on multi-stage pipelines or iterative optimization, leading to high inference latency, failing to meet real-time requirements, and susceptibility to error accumulation. To address these limitations, we propose THO, an end-to-end Spatial-Temporal Transformer that predicts human motion and coordinated object motion in a forward fashion from the given video and 3D template. THO achieves this by leveraging spatial-temporal HOI tuple priors. Spatial priors exploit contact-region proximity to infer occluded object features from human cues, while temporal priors capture cross-frame kinematic correlations to refine object representations and enforce physical coherence. Extensive experiments demonstrate that THO operates at an inference speed of 31.5 FPS on a single RTX 4090 GPU, achieving a >600x speedup over prior optimization-based methods while simultaneously improving reconstruction accuracy and temporal consistency. The project page is available at: https://nianheng.github.io/THO-project/ --- ### Creative Convergence or Imitation? Genre-Specific Homogeneity in LLM-Generated Chinese Literature Authors: Yuanchi Ma, Kaize Shi, Hui He, Zhihua Zhang, Zhongxiang Lei, Ziliang Qiu, Renfen Hu, Jiamou Liu Published: 2026-03-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.14430v1 Large Language Models (LLMs) have demonstrated remarkable capabilities in narrative generation. However, they often produce structurally homogenized stories, frequently following repetitive arrangements and combinations of plot events along with stereotypical resolutions. In this paper, we propose a novel theoretical framework for analysis by incorporating Proppian narratology and narrative functions. This framework is used to analyze the composition of narrative texts generated by LLMs to uncover their underlying narrative logic. Taking Chinese web literature as our research focus, we extend Propp's narrative theory, defining 34 narrative functions suited to modern web narrative structures. We further construct a human-annotated corpus to support the analysis of narrative structures within LLM-generated text. Experiments reveal that the primary reasons for the singular narrative logic and severe homogenization in generated texts are that current LLMs are unable to correctly comprehend the meanings of narrative functions and instead adhere to rigid narrative generation paradigms. --- ### GenState-AI: State-Aware Dataset for Text-to-Video Retrieval on AI-Generated Videos Authors: Minghan Li, Tongna Chen, Tianrui Lv, Yishuai Zhang, Suchao An, Guodong Zhou Published: 2026-03-15 Categories: cs.CV, cs.IR, cs.MM arXiv: https://arxiv.org/abs/2603.14426v1 Existing text-to-video retrieval benchmarks are dominated by real-world footage where much of the semantics can be inferred from a single frame, leaving temporal reasoning and explicit end-state grounding under-evaluated. We introduce GenState-AI, an AI-generated benchmark centered on controlled state transitions, where each query is paired with a main video, a temporal hard negative that differs only in the decisive end-state, and a semantic hard negative with content substitution, enabling fine-grained diagnosis of temporal vs. semantic confusions beyond appearance matching. Using Wan2.2-TI2V-5B, we generate short clips whose meaning depends on precise changes in position, quantity, and object relations, providing controllable evaluation conditions for state-aware retrieval. We evaluate two representative MLLM-based baselines, and observe consistent and interpretable failure patterns: both frequently confuse the main video with the temporal hard negative and over-prefer temporally plausible but end-state-incorrect clips, indicating insufficient grounding to decisive end-state evidence, while being comparatively less sensitive to semantic substitutions. We further introduce triplet-based diagnostic analyses, including relative-order statistics and breakdowns across transition categories, to make temporal vs. semantic failure sources explicit. GenState-AI provides a focused testbed for state-aware, temporally and semantically sensitive text-to-video retrieval, and will be released on huggingface.co. --- ### MBD: A Model-Based Debiasing Framework Across User, Content, and Model Dimensions Authors: Yuantong Li, Lei Yuan, Zhihao Zheng, Weimiao Wu, Songbin Liu, Jeong Min Lee, Ali Selman Aydin, Shaofeng Deng, Junbo Chen, Xinyi Zhang, Hongjing Xia, Sam Fieldman et al. Published: 2026-03-15 Categories: cs.LG, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2603.14422v1 Modern recommendation systems rank candidates by aggregating multiple behavioral signals through a value model. However, many commonly used signals are inherently affected by heterogeneous biases. For example, watch time naturally favors long-form content, loop rate favors short - form content, and comment probability favors videos over images. Such biases introduce two critical issues: (1) value model scores may be systematically misaligned with users' relative preferences - for instance, a seemingly low absolute like probability may represent exceptionally strong interest for a user who rarely engages; and (2) changes in value modeling rules can trigger abrupt and undesirable ecosystem shifts. In this work, we ask a fundamental question: can biased behavioral signals be systematically transformed into unbiased signals, under a user - defined notion of ``unbiasedness'', that are both personalized and adaptive? We propose a general, model-based debiasing (MBD) framework that addresses this challenge by augmenting it with distributional modeling. By conditioning on a flexible subset of features (partial feature set), we explicitly estimate the contextual mean and variance of the engagement distribution for arbitrary cohorts (e.g., specific video lengths or user regions) directly alongside the main prediction. This integration allows the framework to convert biased raw signals into unbiased representations, enabling the construction of higher-level, calibrated signals (such as percentiles or z - scores) suitable for the value model. Importantly, the definition of unbiasedness is flexible and controllable, allowing the system to adapt to different personalization objectives and modeling preferences. Crucially, this is implemented as a lightweight, built-in branch of the existing MTML ranking model, requiring no separate serving infrastructure. --- ### Data Darwinism Part II: DataEvolve -- AI can Autonomously Evolve Pretraining Data Curation Authors: Tiantian Mi, Dongming Shan, Zhen Huang, Yiwei Qin, Muhang Xie, Yuxuan Qiao, Yixiu Liu, Chenyang Zhou, Pengfei Liu Published: 2026-03-15 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.14420v1 Data Darwinism (Part I) established a ten-level hierarchy for data processing, showing that stronger processing can unlock greater data value. However, that work relied on manually designed strategies for a single category. Modern pretraining corpora comprise hundreds of heterogeneous categories spanning domains and content types, each demanding specialized treatment. At this scale, manual strategy design becomes prohibitive. This raises a key question: can strategies evolve in an automated way? We introduce DataEvolve, a framework that enables strategies to evolve through iterative optimization rather than manual design. For each data category, DataEvolve operates in a closed evolutionary loop: it identifies quality issues, generates candidate strategies, executes them on sampled data, evaluates results, and refines approaches across generations. The process accumulates knowledge through an experience pool of discovered issues and a strategy pool tracking performance across iterations. Applied to 8 categories spanning 672B tokens from Nemotron-CC, DataEvolve produces Darwin-CC, a 504B-token dataset with strategies evolved through 30 iterations per category. Training 3B models on 500B tokens, Darwin-CC outperforms raw data (+3.96 points) and achieves a 44.13 average score across 18 benchmarks, surpassing DCLM, Ultra-FineWeb, and FineWeb-Edu, with strong gains on knowledge-intensive tasks such as MMLU. Analysis shows evolved strategies converge on cleaning-focused approaches: targeted noise removal and format normalization with domain-aware preservation, echoing the L4 (Generative Refinement) principles from Part I. Ablation studies confirm iterative evolution is essential: optimized strategies outperform suboptimal ones by 2.93 points, establishing evolutionary strategy design as feasible and necessary for pretraining-scale data curation. --- ### Deep EM with Hierarchical Latent Label Modelling for Multi-Site Prostate Lesion Segmentation Authors: Wen Yan, Yipei Wang, Shiqi Huang, Natasha Thorley, Mark Emberton, Vasilis Stavrinides, Yipeng Hu, Dean Barratt Published: 2026-03-15 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.14418v2 Label variability is a major challenge for prostate lesion segmentation. In multi-site datasets, annotations often reflect centre-specific contouring protocols, causing segmentation networks to overfit to local styles and generalise poorly to unseen sites in inference. We treat each observed annotation as a noisy observation of an underlying latent 'clean' lesion mask, and propose a hierarchical expectation-maximisation (HierEM) framework that alternates between: (1) inferring a voxel-wise posterior distribution over the latent mask, and (2) training a CNN using this posterior as a soft target and estimate site-specific sensitivity and specificity under a hierarchical prior. This hierarchical prior decomposes label-quality into a global mean with site- and case-level deviations, reducing site-specific bias by penalising the likelihood term contributed only by site deviations. Experiments on three cohorts demonstrate that the proposed hierarchical EM framework enhances cross-site generalisation compared to state-of-the-art methods. For pooled-dataset evaluation, the per-site mean DSC ranges from 29.50% to 39.69%; for leave-one-site-out generalisation, it ranges from 27.91% to 32.67%, yielding statistically significant improvements over comparison methods (p<0.039). The method also produces interpretable per-site latent label-quality estimates (sensitivity alpha ranges from 31.5% to 47.3% at specificity beta approximates 0.99), supporting post-hoc analyses of cross-site annotation variability. These results indicate that explicitly modelling site-dependent annotation can improve cross-site generalisation. --- ### Questionnaire Responses Do not Capture the Safety of AI Agents Authors: Max Hellrigel-Holderbaum, Edward James Young Published: 2026-03-15 Categories: cs.CY, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2603.14417v1 As AI systems advance in capabilities, measuring their safety and alignment to human values is becoming paramount. A fast-growing field of AI research is devoted to developing such assessments. However, most current advances therein may be ill-suited for assessing AI systems across real-world deployments. Standard methods prompt large language models (LLMs) in a questionnaire-style to describe their values or behavior in hypothetical scenarios. By focusing on unaugmented LLMs, they fall short of evaluating AI agents, which could actually perform relevant behaviors, hence posing much greater risks. LLMs' engagement with scenarios described by questionnaire-style prompts differs starkly from that of agents based on the same LLMs, as reflected in divergences in the inputs, possible actions, environmental interactions, and internal processing. As such, LLMs' responses to scenario descriptions are unlikely to be representative of the corresponding LLM agents' behavior. We further contend that such assessments make strong assumptions concerning the ability and tendency of LLMs to report accurately about their counterfactual behavior. This makes them inadequate to assess risks from AI systems in real-world contexts as they lack construct validity. We then argue that a structurally identical issue holds for current AI alignment approaches. Lastly, we discuss improving safety assessments and alignment training by taking these shortcomings to heart. --- ### Histo-MExNet: A Unified Framework for Real-World, Cross-Magnification, and Trustworthy Breast Cancer Histopathology Authors: Enam Ahmed Taufika, Md Ahasanul Arafatha, Abhijit Kumar Ghoshb, Md. Tanzim Rezab, Md Ashad Alamc Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14416v1 Accurate and reliable histopathological image classification is essential for breast cancer diagnosis. However, many deep learning models remain sensitive to magnification variability and lack interpretability. To address these challenges, we propose Histo-MExNet, a unified framework designed for scaleinvariant and uncertainty-aware classification. The model integrates DenseNet, ConvNeXt, and EfficientNet backbones within a gated multi-expert architecture, incorporates a prototype learning module for example-driven interpretability, and applies physics-informed regularization to enforce morphology preservation and spatial coherence during feature learning. Monte Carlo Dropout is used to quantify predictive uncertainty. On the BreaKHis dataset, Histo-MExNet achieves 96.97% accuracy under multi-magnification training and demonstrates improved generalization to unseen magnification levels compared to single-expert models, while uncertainty estimation helps identify out-of-distribution samples and reduce overconfident errors, supporting a balanced combination of accuracy, robustness, and interpretability for clinical decision support. --- ### G-ZAP: A Generalizable Zero-Shot Framework for Arbitrary-Scale Pansharpening Authors: Zhiqi Yang, Shan Yin, Jingze Liang, Liang-Jian Deng Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14412v2 Pansharpening aims to fuse a high-resolution panchromatic (PAN) image and a low-resolution multispectral (LRMS) image to produce a high-resolution multispectral (HRMS) image. Recent deep models have achieved strong performance, yet they typically rely on large-scale pretraining and often generalize poorly to unseen real-world image pairs. Prior zero-shot approaches improve real-scene generalization but require per-image optimization, hindering weight reuse, and the above methods are usually limited to a fixed scale. To address this issue, we propose G-ZAP, a generalizable zero-shot framework for arbitrary-scale pansharpening, designed to handle cross-resolution, cross-scene, and cross-sensor generalization. G-ZAP adopts a feature-based implicit neural representation (INR) fusion network as the backbone and introduces a multi-scale, semi-supervised training scheme to enable robust generalization. Extensive experiments on multiple real-world datasets show that G-ZAP achieves state-of-the-art results under PAN-scale fusion in both visual quality and quantitative metrics. Notably, G-ZAP supports weight reuse across image pairs while maintaining competitiveness with per-pair retraining, demonstrating strong potential for efficient real-world deployment. --- ### BiT-MCTS: A Theme-based Bidirectional MCTS Approach to Chinese Fiction Generation Authors: Zhaoyi Li, Xu Zhang, Xiaojun Wan Published: 2026-03-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.14410v3 Generating long-form linear fiction from open-ended themes remains a major challenge for large language models, which frequently fail to guarantee global structure and narrative diversity when using premise-based or linear outlining approaches. We present BiT-MCTS, a theme-driven framework that operationalizes a "climax-first, bidirectional expansion" strategy motivated by Freytag's Pyramid. Given a theme, our method extracts a core dramatic conflict and generates an explicit climax, then employs a bidirectional Monte Carlo Tree Search (MCTS) to expand the plot backward (rising action, exposition) and forward (falling action, resolution) to produce a structured outline. A final generation stage realizes a complete narrative from the refined outline. We construct a Chinese theme corpus for evaluation and conduct extensive experiments across three contemporary LLM backbones. Results show that BiT-MCTS improves narrative coherence, plot structure, and thematic depth relative to strong baselines, while enabling substantially longer, more coherent stories according to automatic metrics and human judgments. --- ### PGcGAN: Pathological Gait-Conditioned GAN for Human Gait Synthesis Authors: Mritula Chandrasekaran, Sanket Kachole, Jarek Francik, Dimitrios Makris Published: 2026-03-15 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.14409v1 Pathological gait analysis is constrained by limited and variable clinical datasets, which restrict the modeling of diverse gait impairments. To address this challenge, we propose a Pathological Gait-conditioned Generative Adversarial Network (PGcGAN) that synthesises pathology-specific gait sequences directly from observed 3D pose keypoint trajectories data. The framework incorporates one-hot encoded pathology labels within both the generator and discriminator, enabling controlled synthesis across six gait categories. The generator adopts a conditional autoencoder architecture trained with adversarial and reconstruction objectives to preserve structural and temporal gait characteristics. Experiments on the Pathological Gait Dataset demonstrate strong alignment between real and synthetic sequences through PCA and t-SNE analyses, visual kinematic inspection, and downstream classification tasks. Augmenting real data with synthetic sequences improved pathological gait recognition across GRU, LSTM, and CNN models, indicating that pathology-conditioned gait synthesis can effectively support data augmentation in pathological gait analysis. --- ### Towards One-for-All Anomaly Detection for Tabular Data Authors: Shiyuan Li, Yixin Liu, Yu Zheng, Xiaofeng Cao, Shirui Pan, Heng Tao Shen Published: 2026-03-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.14407v2 Tabular anomaly detection (TAD) aims to identify samples that deviate from the majority in tabular data and is critical in many real-world applications. However, existing methods follow a ``one model for one dataset (OFO)'' paradigm, which relies on dataset-specific training and thus incurs high computational cost and yields limited generalization to unseen domains. To address these limitations, we propose OFA-TAD, a generalist one-for-all (OFA) TAD framework that only requires one-time training on multiple source datasets and can generalize to unseen datasets from diverse domains on-the-fly. To realize one-for-all tabular anomaly detection, OFA-TAD extracts neighbor-distance patterns as transferable cues, and introduces multi-view neighbor-distance representations from multiple transformation-induced metric spaces to mitigate the transformation sensitivity of distance profiles. To adaptively combine multi-view distance evidence, a Mixture-of-Experts (MoE) scoring network is employed for view-specific anomaly scoring and entropy-regularized gated fusion, with a multi-strategy anomaly synthesis mechanism to support training under the one-class constraint. Extensive experiments on 34 datasets from 14 domains demonstrate that OFA-TAD achieves superior anomaly detection performance and strong cross-domain generalizability under the strict OFA setting. The source code is available at https://github.com/Shiy-Li/OFA-TAD. --- ### Graph-Based Deep Learning for Intelligent Detection of Energy Losses, Theft, and Operational Inefficiencies in Oil & Gas Production Networks Authors: AbdulQoyum A. Olowookere, Adewale U. Oguntola, Ebenezer. Leke Odekanle Published: 2026-03-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.14406v1 Early detection of energy losses, theft, and operational inefficiencies remains a critical challenge in oil and gas production systems due to complex interdependencies among wells and facilities, evolving operating conditions, and limited labeled anomaly data. Traditional machine learning approaches often treat production units independently and struggle under temporal distribution shifts. This study proposes a spatiotemporal graph-based deep learning framework for anomaly detection in oil and gas production networks. The production system is modeled as a hierarchical graph of wells, facilities, and fields, with additional peer connections among wells sharing common infrastructure. Weakly supervised anomaly labels are derived from physically informed heuristics based on production, pressure, and flow behavior. Temporal dynamics are captured through sequence modeling, while relational dependencies are learned using a Temporal Graph Attention Network. Under time-based evaluation, the proposed model achieves an ROC-AUC of about 0.98 and anomaly recall above 0.93, demonstrating improved robustness and practical potential for proactive monitoring in real-world energy operations. --- ### ES-Merging: Biological MLLM Merging via Embedding Space Signals Authors: Wonbin Lee, Dongki Kim, Sung Ju Hwang Published: 2026-03-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.14405v2 Biological multimodal large language models (MLLMs) have emerged as powerful foundation models for scientific discovery. However, existing models are specialized to a single modality, limiting their ability to solve inherently cross-modal scientific problems. While model merging is an efficient method to combine the different modalities into a unified MLLM, existing methods rely on input-agnostic parameter space heuristics that fail to faithfully capture modality specialization. To overcome this limitation, we propose the Embedding-Signal-based MLLM Merging (ES-Merging), a framework that estimates merging coefficients from embedding space signals, moving the merging paradigm from the parameter signals to the embedding signals. ES-Merging exploits coarse-grained and fine-grained signals from embedding space to estimate the layer-wise and element-wise merging coefficients, respectively, which are jointly combined for complementary coefficient estimation. Through extensive experiments, we demonstrate that ES-Merging outperforms existing merging methods not only on the cross-modal reasoning but also on the single-modal knowledge preservation, establishing that embedding space signals provide a principled and effective foundation for MLLM merging. --- ### OCRA: Object-Centric Learning with 3D and Tactile Priors for Human-to-Robot Action Transfer Authors: Kuanning Wang, Ke Fan, Yuqian Fu, Siyu Lin, Hu Luo, Daniel Seita, Yanwei Fu, Yu-Gang Jiang, Xiangyang Xue Published: 2026-03-15 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2603.14401v1 We present OCRA, an Object-Centric framework for video-based human-to-Robot Action transfer that learns directly from human demonstration videos to enable robust manipulation. Object-centric learning emphasizes task-relevant objects and their interactions while filtering out irrelevant background, providing a natural and scalable way to teach robots. OCRA leverages multi-view RGB videos, the state-of-the-art 3D foundation model VGGT, and advanced detection and segmentation models to reconstruct object-centric 3D point clouds, capturing rich interactions between objects. To handle properties not easily perceived by vision alone, we incorporate tactile priors via a large-scale dataset of over one million tactile images. These 3D and tactile priors are fused through a multimodal module (ResFiLM) and fed into a Diffusion Policy to generate robust manipulation actions. Extensive experiments on both vision-only and visuo-tactile tasks show that OCRA significantly outperforms existing baselines and ablations, demonstrating its effectiveness for learning from human demonstration videos. --- ### Extending Minimal Pairs with Ordinal Surprisal Curves and Entropy Across Applied Domains Authors: Andrew Katz Published: 2026-03-15 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.14400v1 The minimal pairs paradigm of comparing model probabilities for contrasting completions has proven useful for evaluating linguistic knowledge in language models, yet its application has largely been confined to binary grammaticality judgments over syntactic phenomena. Additionally, standard prompting-based evaluation requires expensive text generation, may elicit post-hoc rationalizations rather than model judgments, and discards information about model uncertainty. We address both limitations by extending surprisal-based evaluation from binary grammaticality contrasts to ordinal-scaled classification and scoring tasks across multiple domains. Rather than asking models to generate answers, we measure the information-theoretic "surprise" (negative log probability) they assign to each position on rating scales (e.g., 1-5 or 1-9), yielding full surprisal curves that reveal both the model's preferred response and its uncertainty via entropy. We explore this framework across four domains: social-ecological-technological systems classification, causal statement identification (binary and scaled), figurative language detection, and deductive qualitative coding. Across these domains, surprisal curves produce interpretable classification signals with clear minima near expected ordinal scale positions, and entropy over the completion tended to distinguish genuinely ambiguous items from easier items. --- ### WestWorld: A Knowledge-Encoded Scalable Trajectory World Model for Diverse Robotic Systems Authors: Yuchen Wang, Jiangtao Kong, Sizhe Wei, Xiaochang Li, Haohong Lin, Hongjue Zhao, Tianyi Zhou, Lu Gan, Huajie Shao Published: 2026-03-15 Categories: cs.LG, cs.RO arXiv: https://arxiv.org/abs/2603.14392v2 Trajectory world models play a crucial role in robotic dynamics learning, planning, and control. While recent works have explored trajectory world models for diverse robotic systems, they struggle to scale to a large number of distinct system dynamics and overlook domain knowledge of physical structures. To address these limitations, we introduce WestWorld, a knoWledge-Encoded Scalable Trajectory World model for diverse robotic systems. To tackle the scalability challenge, we propose a novel system-aware Mixture-of-Experts (Sys-MoE) that dynamically combines and routes specialized experts for different robotic systems via a learnable system embedding. To further enhance zero-shot generalization, we incorporate domain knowledge of robot physical structures by introducing a structural embedding that aligns trajectory representations with morphological information. After pretraining on 89 complex environments spanning diverse morphologies across both simulation and real-world settings, WestWorld achieves significant improvements over competitive baselines in zero- and few-shot trajectory prediction. Additionally, it shows strong scalability across a wide range of robotic environments and significantly improves performance on downstream model-based control for different robots. Finally, we deploy our model on a real-world Unitree Go1, where it demonstrates stable locomotion performance. The code is available at https://github.com/511205787/WestWorld. --- ### From $\log π$ to $π$: Taming Divergence in Soft Clipping via Bilateral Decoupled Decay of Probability Gradient Weight Authors: Xiaoliang Fu, Jiaye Lin, Yangyi Fang, Chaowen Hu, Cong Qin, Zekai Shao, Binbin Zheng, Lu Pan, Ke Zeng Published: 2026-03-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.14389v2 Reinforcement Learning with Verifiable Rewards (RLVR) has catalyzed a leap in Large Language Model (LLM) reasoning, yet its optimization dynamics remain fragile. Standard algorithms like GRPO enforce stability via "hard clipping", which inadvertently stifles exploration by discarding gradients of tokens outside the trust region. While recent "soft clipping" methods attempt to recover these gradients, they suffer from a critical challenge: relying on log-probability gradient ($\nabla_θ\log π_θ$) yields divergent weights as probabilities vanish, destabilizing LLM training. We rethink this convention by establishing probability gradient ($\nabla_θπ_θ$) as the superior optimization primitive. Accordingly, we propose Decoupled Gradient Policy Optimization (DGPO), which employs a decoupled decay mechanism based on importance sampling ratios. By applying asymmetric, continuous decay to boundary tokens, DGPO resolves the conflict between stability and sustained exploration. Extensive experiments across DeepSeek-R1-Distill-Qwen series models (1.5B/7B/14B) demonstrate that DGPO consistently outperforms strong baselines on various mathematical benchmarks, offering a robust and scalable solution for RLVR. Our code and implementation are available at: https://github.com/FlyTune/DGPO-RL. --- ### Label Noise Cleaning for Supervised Classification via Bernoulli Random Sampling Authors: Yuxin Liu, Xiong Jin, Yang Han Published: 2026-03-15 Categories: stat.ME, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2603.14387v1 Label noise - incorrect labels assigned to observations - can substantially degrade the performance of supervised classifiers. This paper proposes a label noise cleaning method based on Bernoulli random sampling. We show that the mean label noise levels of subsets generated by Bernoulli random sampling containing a given observation are identically distributed for all clean observations, and identically distributed, with a different distribution, for all noisy observations. Although the mean label noise levels are not independent across observations, by introducing an independent coupling we further prove that they converge to a mixture of two well-separated distributions corresponding to clean and noisy observations. By establishing a linear model between cross-validated classification errors and label noise levels, we are able to approximate this mixture distribution and thereby separate clean and noisy observations without any prior label information. The proposed method is classifier-agnostic, theoretically justified, and demonstrates strong performance on both simulated and real datasets. --- ### StAR: Segment Anything Reasoner Authors: Seokju Yun, Dongheon Lee, Noori Bae, Jaesung Jun, Chanseul Cho, Youngmin Ro Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14382v2 As AI systems are being integrated more rapidly into diverse and complex real-world environments, the ability to perform holistic reasoning over an implicit query and an image to localize a target is becoming increasingly important. However, recent reasoning segmentation methods fail to sufficiently elicit the visual reasoning capabilities of the base mode. In this work, we present Segment Anything Reasoner (StAR), a comprehensive framework that refines the design space from multiple perspectives-including parameter-tuning scheme, reward functions, learning strategies and answer format-and achieves substantial improvements over recent baselines. In addition, for the first time, we successfully introduce parallel test-time scaling to the segmentation task, pushing the performance boundary even further. To extend the scope and depth of reasoning covered by existing benchmark, we also construct the ReasonSeg-X, which compactly defines reasoning types and includes samples that require deeper reasoning. Leveraging this dataset, we train StAR with a rollout-expanded selective-tuning approach to activate the base model's latent reasoning capabilities, and establish a rigorous benchmark for systematic, fine-grained evaluation of advanced methods. With only 5k training samples, StAR achieves significant gains over its base counterparts across extensive benchmarks, demonstrating that our method effectively brings dormant reasoning competence to the surface. --- ### SPARQ: Spiking Early-Exit Neural Networks for Energy-Efficient Edge AI Authors: Parth Patne, Mahdi Taheri, Ali Mahani, Maksim Jenihhin, Reza Mahani, Christian Herglotz Published: 2026-03-15 Categories: cs.LG, cs.AI, cs.AR arXiv: https://arxiv.org/abs/2603.14380v1 Spiking neural networks (SNNs) offer inherent energy efficiency due to their event-driven computation model, making them promising for edge AI deployment. However, their practical adoption is limited by the computational overhead of deep architectures and the absence of input-adaptive control. This work presents SPARQ, a unified framework that integrates spiking computation, quantization-aware training, and reinforcement learning-guided early exits for efficient and adaptive inference. Evaluations across MLP, LeNet, and AlexNet architectures demonstrated that the proposed Quantised Dynamic SNNs (QDSNN) consistently outperform conventional SNNs and QSNNs, achieving up to 5.15% higher accuracy over QSNNs, over 330 times lower system energy compared to baseline SNNs, and over 90 percent fewer synaptic operations across different datasets. These results validate SPARQ as a hardware-friendly, energy-efficient solution for real-time AI at the edge. --- ### LoCAtion: Long-time Collaborative Attention Framework for High Dynamic Range Video Reconstruction Authors: Qianyu Zhang, Bolun Zheng, Lingyu Zhu, Aiai Huang, Zongpeng Li, Shiqi Wang Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14377v2 Prevailing High Dynamic Range (HDR) video reconstruction methods are fundamentally trapped in a fragile alignment-and-fusion paradigm. While explicit spatial alignment can successfully recover fine details in controlled environments, it becomes a severe bottleneck in unconstrained dynamic scenes. By forcing rigid alignment across unpredictable motions and varying exposures, these methods inevitably translate registration errors into severe ghosting artifacts and temporal flickering. In this paper, we rethink this conventional prerequisite. Recognizing that explicit alignment is inherently vulnerable to real-world complexities, we propose LoCAtion, a Long-time Collaborative Attention framework that reformulates HDR video generation from a fragile spatial warping task into a robust, alignment-free collaborative feature routing problem. Guided by this new formulation, our architecture explicitly decouples the highly entangled reconstruction task. Rather than struggling to rigidly warp neighboring frames, we anchor the scene on a continuous medium-exposure backbone and utilize collaborative attention to dynamically harvest and inject reliable irradiance cues from unaligned exposures. Furthermore, we introduce a learned global sequence solver. By leveraging bidirectional context and long-range temporal modeling, it propagates corrective signals and structural features across the entire sequence, inherently enforcing whole-video coherence and eliminating jitter. Extensive experiments demonstrate that LoCAtion achieves state-of-the-art visual quality and temporal stability, offering a highly competitive balance between accuracy and computational efficiency. --- ### The Pulse of Motion: Measuring Physical Frame Rate from Visual Dynamics Authors: Xiangbo Gao, Mingyang Wu, Siyuan Yang, Jiongze Yu, Pardis Taghavi, Fangzhou Lin, Zhengzhong Tu Published: 2026-03-15 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.14375v2 While recent generative video models have achieved remarkable visual realism and are being explored as world models, true physical simulation requires mastering both space and time. Current models can produce visually smooth kinematics, yet they lack a reliable internal motion pulse to ground these motions in a consistent, real-world time scale. This temporal ambiguity stems from the common practice of indiscriminately training on videos with vastly different real-world speeds, forcing them into standardized frame rates. This leads to what we term chronometric hallucination: generated sequences exhibit ambiguous, unstable, and uncontrollable physical motion speeds. To address this, we propose Visual Chronometer, a predictor that recovers the Physical Frames Per Second (PhyFPS) directly from the visual dynamics of an input video. Trained via controlled temporal resampling, our method estimates the true temporal scale implied by the motion itself, bypassing unreliable metadata. To systematically quantify this issue, we establish two benchmarks, PhyFPS-Bench-Real and PhyFPS-Bench-Gen. Our evaluations reveal a harsh reality: state-of-the-art video generators suffer from severe PhyFPS misalignment and temporal instability. Finally, we demonstrate that applying PhyFPS corrections significantly improves the human-perceived naturalness of AI-generated videos. Our project page is https://xiangbogaobarry.github.io/Visual_Chronometer/. --- ### Content Creation with Spillovers: An Incentive Design Approach Authors: Sagi Ohayon, Boaz Taitler, Omer Ben-Porat Published: 2026-03-15 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.14372v3 The rise of AI amplifies the economic phenomenon of \emph{positive spillovers}: when creators contribute content that can be reused and adapted by LLMs, one creator's effort may improve the content quality of others through recombination. While such spillovers can improve content quality, they also reshape incentives, as creators may reduce effort when they can benefit from others' contributions. We introduce the \emph{Content Creation with Spillovers} (CCS) model, a game between a platform and strategic creators. In this game, each creator chooses an effort level, and qualities are jointly determined by all creators' efforts. The platform contracts on qualities while seeking to maximize social welfare. We show that standard contest mechanisms can be unstable, and propose a parametrized family of \emph{Provisional Allocation} mechanisms that guarantee a Pareto-dominant equilibrium. Although optimizing welfare within this family is hard to approximate in general, we develop approximation algorithms whose guarantees hold either deterministically across structured spillover classes or with high probability under random instances. --- ### OxyGen: Unified KV Cache Management for VLA Inference under Multi-Task Parallelism Authors: Xiangyu Li, Huaizhi Tang, Xin Ding, Weijun Wang, Ting Cao, Yunxin Liu Published: 2026-03-15 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2603.14371v2 Embodied AI agents increasingly require parallel execution of multiple tasks, such as manipulation, conversation, and memory construction, from shared observations under distinct time constraints. Recent Mixture-of-Transformers (MoT) Vision-Language-Action Models (VLAs) architecturally support such heterogeneous outputs, yet existing inference systems fail to achieve efficient multi-task parallelism for on-device deployment because of redundant computation and resource contention. We identify isolated KV cache management as the root cause. To address this, we propose unified KV cache management, an inference design that treats the KV cache as a first-class shared resource across tasks and over time. This abstraction enables two key optimizations: cross-task KV sharing eliminates redundant prefill of shared observations, while cross-frame continuous batching decouples variable-length language decoding from fixed-rate action generation across control cycles. We implement this design for $π_{0.5}$, a popular MoT VLA, and evaluate it on both NVIDIA GeForce RTX 4090 and Jetson AGX Thor, two representative platforms for on-device VLA inference. OxyGen achieves up to 3.7$\times$ speedup over isolated execution, delivering over 200 tokens/s language throughput and 70 Hz action frequency simultaneously without degrading action quality, and we further validate the gains on a real humanoid robot with on-board Jetson AGX Thor. --- ### From Specification to Architecture: A Theory Compiler for Knowledge-Guided Machine Learning Authors: Asela Hevapathige, Yu Xia, Sachith Seneviratne, Saman Halgamuge Published: 2026-03-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.14369v1 Theory-guided machine learning has demonstrated that including authentic domain knowledge directly into model design improves performance, sample efficiency and out-of-distribution generalisation. Yet the process by which a formal domain theory is translated into architectural constraints remains entirely manual, specific to each domain formalism, and devoid of any formal correctness guarantee. This translation is non-transferable between domains, not verified, and does not scale. We propose the Theory Compiler: a system that accepts a typed, machine-readable domain theory as input and automatically produces an architecture whose function space is provably constrained to be consistent with that theory by construction, not by regularisation. We identify three foundational open problems whose resolution defines our research agenda: (1) designing a universal theory formalisation language with decidable type-checking; (2) constructing a compositionally correct compilation algorithm from theory primitives to architectural modules; and (3) establishing soundness and completeness criteria for formal verification. We further conjecture that compiled architectures match or exceed manually-designed counterparts in generalisation performance while requiring substantially less training data, a claim we ground in classical statistical learning theory. We argue that recent advances in formal machine learning theory, large language models, and the growth of an interdisciplinary research community have made this paradigm achievable for the first time. --- ### HomeGuard: VLM-based Embodied Safeguard for Identifying Contextual Risk in Household Task Authors: Xiaoya Lu, Yijin Zhou, Zeren Chen, Ruocheng Wang, Bingrui Sima, Enshen Zhou, Lu Sheng, Dongrui Liu, Jing Shao Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14367v1 Vision-Language Models (VLMs) empower embodied agents to execute complex instructions, yet they remain vulnerable to contextual safety risks where benign commands become hazardous due to subtle environmental states. Existing safeguards often prove inadequate. Rule-based methods lack scalability in object-dense scenes, whereas model-based approaches relying on prompt engineering suffer from unfocused perception, resulting in missed risks or hallucinations. To address this, we propose an architecture-agnostic safeguard featuring Context-Guided Chain-of-Thought (CG-CoT). This mechanism decomposes risk assessment into active perception that sequentially anchors attention to interaction targets and relevant spatial neighborhoods, followed by semantic judgment based on this visual evidence. We support this approach with a curated grounding dataset and a two-stage training strategy utilizing Reinforcement Fine-Tuning (RFT) with process rewards to enforce precise intermediate grounding. Experiments demonstrate that our model HomeGuard significantly enhances safety, improving risk match rates by over 30% compared to base models while reducing oversafety. Beyond hazard detection, the generated visual anchors serve as actionable spatial constraints for downstream planners, facilitating explicit collision avoidance and safety trajectory generation. Code and data are released under https://github.com/AI45Lab/HomeGuard --- ### Representation Alignment for Just Image Transformers is not Easier than You Think Authors: Jaeyo Shin, Jiwook Kim, Hyunjung Shim Published: 2026-03-15 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2603.14366v1 Representation Alignment (REPA) has emerged as a simple way to accelerate Diffusion Transformers training in latent space. At the same time, pixel-space diffusion transformers such as Just image Transformers (JiT) have attracted growing attention because they remove a dependency on a pretrained tokenizer, and then avoid the reconstruction bottleneck of latent diffusion. This paper shows that the REPA can fail for JiT. REPA yields worse FID for JiT as training proceeds and collapses diversity on image subsets that are tightly clustered in the representation space of pretrained semantic encoder on ImageNet. We trace the failure to an information asymmetry: denoising occurs in the high dimensional image space, while the semantic target is strongly compressed, making direct regression a shortcut objective. We propose PixelREPA, which transforms the alignment target and constrains alignment with a Masked Transformer Adapter that combines a shallow transformer adapter with partial token masking. PixelREPA improves both training convergence and final quality. PixelREPA reduces FID from 3.66 to 3.17 for JiT-B$/16$ and improves Inception Score (IS) from 275.1 to 284.6 on ImageNet $256 \times 256$, while achieving $> 2\times$ faster convergence. Finally, PixelREPA-H$/16$ achieves FID$=1.81$ and IS$=317.2$. Our code is available at https://github.com/kaist-cvml/PixelREPA. --- ### AerialVLA: A Vision-Language-Action Model for UAV Navigation via Minimalist End-to-End Control Authors: Peng Xu, Zhengnan Deng, Jiayan Deng, Zonghua Gu, Shaohua Wan Published: 2026-03-15 Categories: cs.CV, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2603.14363v1 Vision-Language Navigation (VLN) for Unmanned Aerial Vehicles (UAVs) demands complex visual interpretation and continuous control in dynamic 3D environments. Existing hierarchical approaches rely on dense oracle guidance or auxiliary object detectors, creating semantic gaps and limiting genuine autonomy. We propose AerialVLA, a minimalist end-to-end Vision-Language-Action framework mapping raw visual observations and fuzzy linguistic instructions directly to continuous physical control signals. First, we introduce a streamlined dual-view perception strategy that reduces visual redundancy while preserving essential cues for forward navigation and precise grounding, which additionally facilitates future simulation-to-reality transfer. To reclaim genuine autonomy, we deploy a fuzzy directional prompting mechanism derived solely from onboard sensors, completely eliminating the dependency on dense oracle guidance. Ultimately, we formulate a unified control space that integrates continuous 3-Degree-of-Freedom (3-DoF) kinematic commands with an intrinsic landing signal, freeing the agent from external object detectors for precision landing. Extensive experiments on the TravelUAV benchmark demonstrate that AerialVLA achieves state-of-the-art performance in seen environments. Furthermore, it exhibits superior generalization in unseen scenarios by achieving nearly three times the success rate of leading baselines, validating that a minimalist, autonomy-centric paradigm captures more robust visual-motor representations than complex modular systems. --- ### BROTHER: Behavioral Recognition Optimized Through Heterogeneous Ensemble Regularization for Ambivalence and Hesitancy Authors: Alexandre Pereira, Bruno Fernandes, Pablo Barros Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14361v1 Recognizing complex behavioral states such as Ambivalence and Hesitancy (A/H) in naturalistic video settings remains a significant challenge in affective computing. Unlike basic facial expressions, A/H manifests as subtle, multimodal conflicts that require deep contextual and temporal understanding. In this paper, we propose a highly regularized, multimodal fusion pipeline to predict A/H at the video level. We extract robust unimodal features from visual, acoustic, and linguistic data, introducing a specialized statistical text modality explicitly designed to capture temporal speech variations and behavioral cues. To identify the most effective representations, we evaluate 15 distinct modality combinations across a committee of machine learning classifiers (MLP, Random Forest, and GBDT), selecting the most well-calibrated models based on validation Binary Cross-Entropy (BCE) loss. Furthermore, to optimally fuse these heterogeneous models without overfitting to the training distribution, we implement a Particle Swarm Optimization (PSO) hard-voting ensemble. The PSO fitness function dynamically incorporates a train-validation gap penalty (lambda) to actively suppress redundant or overfitted classifiers. Our comprehensive evaluation demonstrates that while linguistic features serve as the strongest independent predictor of A/H, our heavily regularized PSO ensemble (lambda = 0.2) effectively harnesses multimodal synergies, achieving a peak Macro F1-score of 0.7465 on the unseen test set. These results emphasize that treating ambivalence and hesitancy as a multimodal conflict, evaluated through an intelligently weighted committee, provides a robust framework for in-the-wild behavioral analysis. --- ### M$^2$RNN: Non-Linear RNNs with Matrix-Valued States for Scalable Language Modeling Authors: Mayank Mishra, Shawn Tan, Ion Stoica, Joseph Gonzalez, Tri Dao Published: 2026-03-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.14360v2 Transformers are highly parallel but are limited to computations in the TC$^0$ complexity class, excluding tasks such as entity tracking and code execution that provably require greater expressive power. Motivated by this limitation, we revisit non-linear Recurrent Neural Networks (RNNs) for language modeling and introduce Matrix-to-Matrix RNN (M$^2$RNN): an architecture with matrix-valued hidden states and expressive non-linear state transitions. We demonstrate that the language modeling performance of non-linear RNNs is limited by their state size, and show how the state size expansion mechanism enables efficient use of tensor cores. Empirically, M$^2$RNN achieves perfect state tracking generalization at sequence lengths not seen during training. These benefits also translate to large-scale language modeling. In hybrid settings that interleave recurrent layers with attention, Hybrid M$^2$RNN outperforms equivalent Gated DeltaNet hybrids by $0.4$-$0.5$ perplexity points on a 7B MoE model, while using $3\times$ smaller state sizes for the recurrent layers. Notably, replacing even a single recurrent layer with M$^2$RNN in an existing hybrid architecture yields accuracy gains comparable to Hybrid M$^2$RNN with minimal impact on training throughput. Further, the Hybrid Gated DeltaNet models with a single M$^2$RNN layer also achieve superior long-context generalization, outperforming state-of-the-art hybrid linear attention architectures by up to $8$ points on LongBench. Together, these results establish non-linear RNN layers as a compelling building block for efficient and scalable language models. --- ### Exposing Long-Tail Safety Failures in Large Language Models through Efficient Diverse Response Sampling Authors: Suvadeep Hajra, Palash Nandi, Tanmoy Chakraborty Published: 2026-03-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.14355v2 Safety tuning through supervised fine-tuning and reinforcement learning from human feedback has substantially improved the robustness of large language models. However, it typically suppresses rather eliminates unsafe behaviors, leaving rare but critical failures hidden in the long tail of the output distribution. While most red-teaming work emphasizes adversarial prompt search, we show that these hidden risks can be systematically exposed through diverse response generation. Specifically, we show that, for a fixed safety-critical prompt, increasing the number and diversity of sampled responses monotonically raises the jailbreak success rate. To efficiently uncover these failures, we propose Progressive Diverse Population Sampling (PDPS). This approach replaces naive, large-scale IID sampling with a multi-stage expansion-and-selection strategy that generates a compact, semantically diverse set of responses at a substantially lower computational cost. Across multiple jailbreak benchmarks and open-source LLMs, PDPS achieves attack success rates comparable to large-scale IID sampling while using only 8%-29% of the computational cost, and outperforms IID sampling and Diverse Beam Search by 26%-40% under limited-response budgets, while uncovering a broader and more semantically diverse range of failure modes. Critically, this diversity translates directly into more effective safety hardening: when integrated into an RLHF-based safety-tuning pipeline, PDPS-generated unsafe responses yield 33% and 41% greater reductions in ASR than those generated by IID sampling and Diverse Beam Search, respectively. Finally, we show that while input-space prompt optimization methods fall short of output-space exploration when used in isolation, combining input-space perturbation with diversity-driven output-space exploration covers a wider range of failure modes more efficiently than either paradigm alone. --- ### Deconfounded Lifelong Learning for Autonomous Driving via Dynamic Knowledge Spaces Authors: Jiayuan Du, Yuebing Song, Yiming Zhao, Xianghui Pan, Jiawei Lian, Yuchu Lu, Liuyi Wang, Chengju Liu, Qijun Chen Published: 2026-03-15 Categories: cs.LG, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2603.14354v3 End-to-End autonomous driving (E2E-AD) systems face challenges in lifelong learning, including catastrophic forgetting, difficulty in knowledge transfer across diverse scenarios, and spurious correlations between unobservable confounders and true driving intents. To address these issues, we propose DeLL, a Deconfounded Lifelong Learning framework that integrates a Dirichlet process mixture model (DPMM) with the front-door adjustment mechanism from causal inference. The DPMM is employed to construct two dynamic knowledge spaces: a trajectory knowledge space for clustering explicit driving behaviors and an implicit feature knowledge space for discovering latent driving abilities. Leveraging the non-parametric Bayesian nature of DPMM, our framework enables adaptive expansion and incremental updating of knowledge without predefining the number of clusters, thereby mitigating catastrophic forgetting. Meanwhile, the front-door adjustment mechanism utilizes the DPMM-derived knowledge as mediators to deconfound spurious correlations, such as those induced by sensor noise or environmental changes, and enhances the causal expressiveness of the learned representations. Additionally, we introduce an evolutionary trajectory decoder that enables non-autoregressive planning. To evaluate the lifelong learning performance of E2E-AD, we propose new evaluation protocols and metrics based on Bench2Drive. Extensive evaluations in the closed-loop CARLA simulator demonstrate that our framework significantly improves adaptability to new driving scenarios and overall driving performance, while effectively retaining previously acquired knowledge. Code: https://github.com/Mooncakebro/DeLL --- ### Refold: Refining Protein Inverse Folding with Efficient Structural Matching and Fusion Authors: Yiran Zhu, Changxi Chi, Hongxin Xiang, Wenjie Du, Xiaoqi Wang, Jun Xia Published: 2026-03-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.14350v1 Protein inverse folding aims to design an amino acid sequence that will fold into a given backbone structure, serving as a central task in protein design. Two main paradigms have been widely explored. Template-based methods exploit database-derived structural priors and can achieve high local precision when close structural neighbors are available, but their dependence on database coverage and match quality often degrades performance on out-of-distribution (OOD) targets. Deep learning approaches, in contrast, learn general structure-to-sequence regularities and usually generalize better to new backbones. However, they struggle to capture fine-grained local structure, which can cause uncertain residue predictions and missed local motifs in ambiguous regions. We introduce Refold, a novel framework that synergistically integrates the strengths of database-derived structural priors and deep learning prediction to enhance inverse folding. Refold obtains structural priors from matched neighbors and fuses them with model predictions to refine residue probabilities. In practice, low-quality neighbors can introduce noise, potentially degrading model performance. We address this issue with a Dynamic Utility Gate that controls prior injection and falls back to the base prediction when the priors are untrustworthy. Comprehensive evaluations on standard benchmarks demonstrate that Refold achieves state-of-the-art native sequence recovery of 0.63 on both CATH 4.2 and CATH 4.3. Also, analysis indicates that Refold delivers larger gains on high-uncertainty regions, reflecting the complementarity between structural priors and deep learning predictions. --- ### Motivation in Large Language Models Authors: Omer Nahum, Asael Sklar, Ariel Goldstein, Roi Reichart Published: 2026-03-15 Categories: cs.CL, cs.CY arXiv: https://arxiv.org/abs/2603.14347v1 Motivation is a central driver of human behavior, shaping decisions, goals, and task performance. As large language models (LLMs) become increasingly aligned with human preferences, we ask whether they exhibit something akin to motivation. We examine whether LLMs "report" varying levels of motivation, how these reports relate to their behavior, and whether external factors can influence them. Our experiments reveal consistent and structured patterns that echo human psychology: self-reported motivation aligns with different behavioral signatures, varies across task types, and can be modulated by external manipulations. These findings demonstrate that motivation is a coherent organizing construct for LLM behavior, systematically linking reports, choices, effort, and performance, and revealing motivational dynamics that resemble those documented in human psychology. This perspective deepens our understanding of model behavior and its connection to human-inspired concepts. --- ### Localizing and Editing Knowledge in Large Audio-Language Models Authors: Sung Kyun Chung, Jiaheng Dong, Qiuchi Hu, Gongping Huang, Hong Jia, Ting Dang Published: 2026-03-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.14343v2 Large Audio-Language Models (LALMs) have shown strong performance in speech understanding, making speech a natural interface for accessing factual information. Yet they are trained on static corpora and may encode incorrect facts. Existing model editing methods localize and update facts in text-only LLMs, but do not account for continuous speech representations, or where knowledge is stored across acoustic or language modules, or their cross-modal module. We construct the first audio benchmark for knowledge localization and editing in LALMs and propose a speech-driven locate-then-edit framework. First, we use speech-aware causal tracing to localize layers and modules that support factual retrieval and then apply editing at identified sites. Experiments show that factual knowledge is jointly encoded in audio and text modules, and that audio editing yields more effective updates than text editing or fine-tuning, enabling fine-grained knowledge control in speech AI systems. --- ### AgroOmni: A Large-Scale Multi-view Agricultural Dataset for Cross-Scale Multimodal Reasoning Authors: Jiarui Zhang, Junqi Hu, Zurong Mai, Yang Liu, Yuhang Chen, Shuohong Lou, Henglian Huang, Hong Cheng, Lingyuan Zhao, Jianxi Huang, Yutong Lu, Haohuan Fu et al. Published: 2026-03-15 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.14342v2 Modern agricultural data is sourced from diverse platforms and spans multiple spatial scales, ranging from ground-level close-up photography to Unmanned Aerial Vehicle (UAV) aerial observation and satellite remote sensing imagery. Accordingly, agricultural multimodal reasoning demands robust cross-scale spatial understanding. However, due to the lack of multi-view agricultural benchmark datasets, existing multimodal large language models (MLLMs) exhibit severe ground-level bias, which leads to scale confusion then semantic collapse in agricultural perception tasks, such as misinterpreting farmland imagery as walls or floors. To address this, we introduce AgroOmni, a large-scale multi-view training corpus with 288K Visual Question Answering pairs covering 56 specialized task categories across 14 task types, designed to capture diverse scales in modern precision agriculture. Built on this dataset, we propose AgroNVILA, which achieves a new state-of-the-art of 62.32% on the AgroMind benchmark (+15.03% over GPT-5.2), effectively mitigating the multi-view cross-scale gap for holistic agricultural understanding. Diagnostic evaluations on AgMMU further reveal an inherent heterogeneity between macro-priors and micro-diagnostics through constrained zero-shot performance. Meanwhile, even minimal fine-tuning leads to a dramatic performance gain of AgroNVILA on AgMMU, strongly demonstrating its generalization capability empowered by AgroOmni. Full training scripts are publicly available at https://anonymous.4open.science/r/AgroOmni-6510. --- ### Generation of Human Comprehensible Access Control Policies from Audit Logs Authors: Gautam Kumar, Ravi Sundaram, Shamik Sural Published: 2026-03-15 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2603.14341v1 Over the years, access control systems have become increasingly more complex, often causing a disconnect between what is envisaged by the stakeholders in decision-making positions and the actual permissions granted as evidenced from access logs. For instance, Attribute-based Access Control (ABAC), which is a flexible yet complex model typically configured by system security officers, can be made understandable to others only when presented at a high level in natural language. Although several algorithms have been proposed in the literature for automatic extraction of ABAC rules from access logs, there is no attempt yet to bridge the semantic gap between the machine-enforceable formal logic and human-centric policy intent. Our work addresses this problem by developing a framework that generates human understandable natural language access control policies from logs. We investigate to what extent the power of Large Language Models (LLMs) can be harnessed to achieve both accuracy and scalability in the process. Named LANTERN (LLM-based ABAC Natural Translation and Explanation for Rule Navigation), we have instantiated the framework as a publicly accessible web based application for reproducibility of our results. --- ### On the Nature of Attention Sink that Shapes Decoding Strategy in Omni-LLMs Authors: Suho Yoo, Youngjoon Jang, Joon Son Chung Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14337v2 The goal of this paper is to strengthen the reasoning of Omnimodal Large Language Models (Omni-LLMs) at inference time, without additional training. These models jointly process video, audio, and text, and given the large number of tokens they consume, how attention is routed across them is central to their behaviour. We focus specifically on attention sinks, tokens that absorb a disproportionate share of attention mass regardless of their semantic content, to understand how this routing unfolds. To this end, we conduct a systematic analysis of sink behaviour in Omni-LLMs. Our analysis yields two key findings: (i) high sink attention does not solely indicate head redundancy, suggesting that sink value representations play additional functional roles; (ii) the sink value vector acts as a shared bias added to every token's output, serving as a global signal that organises the representation as a whole. Building on this, we propose OutRo, which correspondingly aligns non-sink token representations with the sink in feature space, and relaxes the causal mask for sink tokens at an early layer to sharpen this bias before the rest of decoding proceeds. This design enhances the reasoning process without requiring additional forward passes or access to attention maps. Based on extensive experiments, OutRo consistently improves performance on seven video QA benchmarks and demonstrates strong generalisation, while incurring only a 1.1x decoding overhead. --- ### UAVBench and UAVIT-1M: Benchmarking and Enhancing MLLMs for Low-Altitude UAV Vision-Language Understanding Authors: Yang Zhan, Yuan Yuan Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14336v1 Multimodal Large Language Models (MLLMs) have made significant strides in natural images and satellite remote sensing images. However, understanding low-altitude drone scenarios remains a challenge. Existing datasets primarily focus on a few specific low-altitude visual tasks, which cannot fully assess the ability of MLLMs in real-world low-altitude UAV applications. Therefore, we introduce UAVBench, a comprehensive benchmark, and UAVIT-1M, a large-scale instruction tuning dataset, designed to evaluate and improve MLLMs' abilities in low-altitude vision-language tasks. UAVBench comprises 43 test units and 966k high-quality data samples across 10 tasks at the image-level and region-level. UAVIT-1M consists of approximately 1.24 million diverse instructions, covering 789k multi-scene images and about 2,000 types of spatial resolutions with 11 distinct tasks. UAVBench and UAVIT-1M feature pure real-world visual images and rich weather conditions, and involve manual verification to ensure high quality. Our in-depth analysis of 11 state-of-the-art MLLMs using UAVBench reveals that open-source MLLMs cannot generate accurate conversations about low-altitude visual content, lagging behind closed-source MLLMs. Extensive experiments demonstrate that fine-tuning open-source MLLMs on UAVIT-1M significantly addresses this gap. Our contributions pave the way for bridging the gap between current MLLMs and low-altitude UAV real-world application demands. (Project page: https://UAVBench.github.io/) --- ### Data-Driven Physics Embedded Dynamics with Predictive Control and Reinforcement Learning for Quadrupeds Authors: Prakrut Kotecha, Aditya Shirwatkar, Shishir Kolathaya Published: 2026-03-15 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2603.14333v1 State of the art quadrupedal locomotion approaches integrate Model Predictive Control (MPC) with Reinforcement Learning (RL), enabling complex motion capabilities with planning and terrain adaptive behaviors. However, they often face compounding errors over long horizons and have limited interpretability due to the absence of physical inductive biases. We address these issues by integrating Lagrangian Neural Networks (LNNs) into an RL MPC framework, enabling physically consistent dynamics learning. At deployment, our inverse dynamics infinite horizon MPC scheme avoids costly matrix inversions, improving computational efficiency by up to 4x with minimal loss of task performance. We validate our framework through multiple ablations of the proposed LNN and its variants. We show improved sample efficiency, reduced long-horizon error, and faster real time planning compared to unstructured neural dynamics. Lastly, we also test our framework on the Unitree Go1 robot to show real world viability. --- ### AvatarForcing: One-Step Streaming Talking Avatars via Local-Future Sliding-Window Denoising Authors: Liyuan Cui, Wentao Hu, Wenyuan Zhang, Zesong Yang, Fan Shi, Xiaoqiang Liu Published: 2026-03-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.14331v2 Real-time talking avatar generation requires low latency and minute-level temporal stability. Autoregressive (AR) forcing enables streaming inference but suffers from exposure bias, which causes errors to accumulate and become irreversible over long rollouts. In contrast, full-sequence diffusion transformers mitigate drift but remain computationally prohibitive for real-time long-form synthesis. We present AvatarForcing, a one-step streaming diffusion framework that denoises a fixed local-future window with heterogeneous noise levels and emits one clean block per step under constant per-step cost. To stabilize unbounded streams, the method introduces dual-anchor temporal forcing: a style anchor that re-indexes RoPE to maintain a fixed relative position with respect to the active window and applies anchor-audio zero-padding, and a temporal anchor that reuses recently emitted clean blocks to ensure smooth transitions. Real-time one-step inference is enabled by two-stage streaming distillation with offline ODE backfill and distribution matching. Experiments on standard benchmarks and a new 400-video long-form benchmark show strong visual quality and lip synchronization at 34 ms/frame using a 1.3B-parameter student model for realtime streaming. Our page is available at: https://cuiliyuan121.github.io/AvatarForcing/ --- ### ECG-Reasoning-Benchmark: A Benchmark for Evaluating Clinical Reasoning Capabilities in ECG Interpretation Authors: Jungwoo Oh, Hyunseung Chung, Junhee Lee, Min-Gyu Kim, Hangyul Yoon, Ki Seong Lee, Youngchae Lee, Muhan Yeo, Edward Choi Published: 2026-03-15 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2603.14326v1 While Multimodal Large Language Models (MLLMs) show promising performance in automated electrocardiogram interpretation, it remains unclear whether they genuinely perform actual step-by-step reasoning or just rely on superficial visual cues. To investigate this, we introduce \textbf{ECG-Reasoning-Benchmark}, a novel multi-turn evaluation framework comprising over 6,400 samples to systematically assess step-by-step reasoning across 17 core ECG diagnoses. Our comprehensive evaluation of state-of-the-art models reveals a critical failure in executing multi-step logical deduction. Although models possess the medical knowledge to retrieve clinical criteria for a diagnosis, they exhibit near-zero success rates (6% Completion) in maintaining a complete reasoning chain, primarily failing to ground the corresponding ECG findings to the actual visual evidence in the ECG signal. These results demonstrate that current MLLMs bypass actual visual interpretation, exposing a critical flaw in existing training paradigms and underscoring the necessity for robust, reasoning-centric medical AI. The code and data are available at https://github.com/Jwoo5/ecg-reasoning-benchmark. ---