# Research Papers — 2026-02 · bundle page 120 of 210 > Full-text markdown of papers #5951–6000 of 10493 published in 2026-02, newest first. AI-readable mirrors of the full 10493-paper 2026-02 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-02/page/120 [← Previous bundle](https://aipod.dev/papers/archive/2026-02/page/119.md) [Next bundle →](https://aipod.dev/papers/archive/2026-02/page/121.md) ### Egocentric Bias in Vision-Language Models Authors: Maijunxian Wang, Yijiang Li, Bingyang Wang, Tianwei Zhao, Ran Ji, Qingying Gao, Emmy Liu, Hokin Deng, Dezhi Luo Published: 2026-02-10 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.15892v2 Visual perspective taking--inferring how the world appears from another's viewpoint--is foundational to social cognition. We introduce FlipSet, a diagnostic benchmark for Level-2 visual perspective taking (L2 VPT) in vision-language models. The task requires simulating 180-degree rotations of 2D character strings from another agent's perspective, isolating spatial transformation from 3D scene complexity. Evaluating 103 VLMs reveals systematic egocentric bias: the vast majority perform below chance, with roughly three-quarters of errors reproducing the camera viewpoint. Control experiments expose a compositional deficit--models achieve high theory-of-mind accuracy and above-chance mental rotation in isolation, yet fail catastrophically when integration is required. This dissociation indicates that current VLMs lack the mechanisms needed to bind social awareness to spatial operations, suggesting fundamental limitations in model-based spatial reasoning. FlipSet provides a cognitively grounded testbed for diagnosing perspective-taking capabilities in multimodal systems. --- ### Synthesizing the Kill Chain: A Zero-Shot Framework for Target Verification and Tactical Reasoning on the Edge Authors: Jesse Barkley, Abraham George, Amir Barati Farimani Published: 2026-02-10 Categories: cs.CV, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2602.13324v1 Deploying autonomous edge robotics in dynamic military environments is constrained by both scarce domain-specific training data and the computational limits of edge hardware. This paper introduces a hierarchical, zero-shot framework that cascades lightweight object detection with compact Vision-Language Models (VLMs) from the Qwen and Gemma families (4B-12B parameters). Grounding DINO serves as a high-recall, text-promptable region proposer, and frames with high detection confidence are passed to edge-class VLMs for semantic verification. We evaluate this pipeline on 55 high-fidelity synthetic videos from Battlefield 6 across three tasks: false-positive filtering (up to 100% accuracy), damage assessment (up to 97.5%), and fine-grained vehicle classification (55-90%). We further extend the pipeline into an agentic Scout-Commander workflow, achieving 100% correct asset deployment and a 9.8/10 reasoning score (graded by GPT-4o) with sub-75-second latency. A novel "Controlled Input" methodology decouples perception from reasoning, revealing distinct failure phenotypes: Gemma3-12B excels at tactical logic but fails in visual perception, while Gemma3-4B exhibits reasoning collapse even with accurate inputs. These findings validate hierarchical zero-shot architectures for edge autonomy and provide a diagnostic framework for certifying VLM suitability in safety-critical applications. --- ### Contrastive explanations of BDI agents Authors: Michael Winikoff Published: 2026-02-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.13323v1 The ability of autonomous systems to provide explanations is important for supporting transparency and aiding the development of (appropriate) trust. Prior work has defined a mechanism for Belief-Desire-Intention (BDI) agents to be able to answer questions of the form ``why did you do action $X$?''. However, we know that we ask \emph{contrastive} questions (``why did you do $X$ \emph{instead of} $F$?''). We therefore extend previous work to be able to answer such questions. A computational evaluation shows that using contrastive questions yields a significant reduction in explanation length. A human subject evaluation was conducted to assess whether such contrastive answers are preferred, and how well they support trust development and transparency. We found some evidence for contrastive answers being preferred, and some evidence that they led to higher trust, perceived understanding, and confidence in the system's correctness. We also evaluated the benefit of providing explanations at all. Surprisingly, there was not a clear benefit, and in some situations we found evidence that providing a (full) explanation was worse than not providing any explanation. --- ### Diagnostic Benchmarks for Invariant Learning Dynamics: Empirical Validation of the Eidos Architecture Authors: Datorien L. Anderson Published: 2026-02-10 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2602.13322v1 We present the PolyShapes-Ideal (PSI) dataset, a suite of diagnostic benchmarks designed to isolate topological invariance -- the ability to maintain structural identity across affine transformations -- from the textural correlations that dominate standard vision benchmarks. Through three diagnostic probes (polygon classification under noise, zero-shot font transfer from MNIST, and geometric collapse mapping under progressive deformation), we demonstrate that the Eidos architecture achieves >99% accuracy on PSI and 81.67% zero-shot transfer across 30 unseen typefaces without pre-training. These results validate the "Form-First" hypothesis: generalization in structurally constrained architectures is a property of geometric integrity, not statistical scale. --- ### Detecting Jailbreak Attempts in Clinical Training LLMs Through Automated Linguistic Feature Extraction Authors: Tri Nguyen, Huy Hoang Bao Le, Lohith Srikanth Pentapalli, Laurah Turner, Kelly Cohen Published: 2026-02-10 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.13321v1 Detecting jailbreak attempts in clinical training large language models (LLMs) requires accurate modeling of linguistic deviations that signal unsafe or off-task user behavior. Prior work on the 2-Sigma clinical simulation platform showed that manually annotated linguistic features could support jailbreak detection. However, reliance on manual annotation limited both scalability and expressiveness. In this study, we extend this framework by using experts' annotations of four core linguistic features (Professionalism, Medical Relevance, Ethical Behavior, and Contextual Distraction) and training multiple general-domain and medical-domain BERT-based LLM models to predict these features directly from text. The most reliable feature regressor for each dimension was selected and used as the feature extractor in a second layer of classifiers. We evaluate a suite of predictive models, including tree-based, linear, probabilistic, and ensemble methods, to determine jailbreak likelihood from the extracted features. Across cross-validation and held-out evaluations, the system achieves strong overall performance, indicating that LLM-derived linguistic features provide an effective basis for automated jailbreak detection. Error analysis further highlights key limitations in current annotations and feature representations, pointing toward future improvements such as richer annotation schemes, finer-grained feature extraction, and methods that capture the evolving risk of jailbreak behavior over the course of a dialogue. This work demonstrates a scalable and interpretable approach for detecting jailbreak behavior in safety-critical clinical dialogue systems. --- ### Information Fidelity in Tool-Using LLM Agents: A Martingale Analysis of the Model Context Protocol Authors: Flint Xiaofeng Fan, Cheston Tan, Roger Wattenhofer, Yew-Soon Ong Published: 2026-02-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.13320v1 As AI agents powered by large language models (LLMs) increasingly use external tools for high-stakes decisions, a critical reliability question arises: how do errors propagate across sequential tool calls? We introduce the first theoretical framework for analyzing error accumulation in Model Context Protocol (MCP) agents, proving that cumulative distortion exhibits linear growth and high-probability deviations bounded by $O(\sqrt{T})$. This concentration property ensures predictable system behavior and rules out exponential failure modes. We develop a hybrid distortion metric combining discrete fact matching with continuous semantic similarity, then establish martingale concentration bounds on error propagation through sequential tool interactions. Experiments across Qwen2-7B, Llama-3-8B, and Mistral-7B validate our theoretical predictions, showing empirical distortion tracks the linear trend with deviations consistently within $O(\sqrt{T})$ envelopes. Key findings include: semantic weighting reduces distortion by 80\%, and periodic re-grounding approximately every 9 steps suffices for error control. We translate these concentration guarantees into actionable deployment principles for trustworthy agent systems. --- ### Situation Graph Prediction for User Perspective Modeling Authors: Jisung Shin, Daniel Platnick, Marjan Alirezaie, Hossein Rahnama Published: 2026-02-10 Categories: cs.AI, cs.HC arXiv: https://arxiv.org/abs/2602.13319v2 Perspective-aware AI requires modeling evolving internal states---goals, emotions, contexts---not merely preferences. Progress is limited by a data bottleneck: digital footprints are privacy-sensitive and perspective states are rarely labeled. We propose Situation Graph Prediction (SGP), a task that frames user perspective modeling as an inverse inference problem: reconstructing structured, ontology-aligned representations of perspective from observable multimodal artifacts, suitable as long-horizon memory for personal agents. To enable grounding without real labels, we use a structure-first synthetic generation strategy that aligns latent labels and observable traces by design. As a pilot, we construct a dataset and run a diagnostic study using retrieval-augmented in-context learning as a proxy for supervision. In our diagnostic study across three frontier foundation models (GPT-4o, Gemini 2.5 Flash, Claude Sonnet 4), we observe a consistent positive gap between surface-level extraction and latent perspective inference---indicating that latent-state inference is consistently harder than surface extraction under our controlled setting. Results suggest SGP is non-trivial and provide evidence for the structure-first data synthesis strategy. For reproducibility and transparency, our code, data, and supplementary resources are available here: https://github.com/flybits/sg-prediction. --- ### Situation Graph Prediction: Structured Perspective Inference for User Modeling Authors: Jisung Shin, Daniel Platnick, Marjan Alirezaie, Hossein Rahnama Published: 2026-02-10 Categories: cs.AI, cs.HC arXiv: https://arxiv.org/abs/2602.13319v1 Perspective-Aware AI requires modeling evolving internal states--goals, emotions, contexts--not merely preferences. Progress is limited by a data bottleneck: digital footprints are privacy-sensitive and perspective states are rarely labeled. We propose Situation Graph Prediction (SGP), a task that frames perspective modeling as an inverse inference problem: reconstructing structured, ontology-aligned representations of perspective from observable multimodal artifacts. To enable grounding without real labels, we use a structure-first synthetic generation strategy that aligns latent labels and observable traces by design. As a pilot, we construct a dataset and run a diagnostic study using retrieval-augmented in-context learning as a proxy for supervision. In our study with GPT-4o, we observe a gap between surface-level extraction and latent perspective inference--indicating latent-state inference is harder than surface extraction under our controlled setting. Results suggest SGP is non-trivial and provide evidence for the structure-first data synthesis strategy. --- ### DECKBench: Benchmarking Multi-Agent Frameworks for Academic Slide Generation and Editing Authors: Daesik Jang, Morgan Lindsay Heisler, Linzi Xing, Yifei Li, Edward Wang, Ying Xiong, Yong Zhang, Zhenan Fan Published: 2026-02-10 Categories: cs.AI, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2602.13318v1 Automatically generating and iteratively editing academic slide decks requires more than document summarization. It demands faithful content selection, coherent slide organization, layout-aware rendering, and robust multi-turn instruction following. However, existing benchmarks and evaluation protocols do not adequately measure these challenges. To address this gap, we introduce the Deck Edits and Compliance Kit Benchmark (DECKBench), an evaluation framework for multi-agent slide generation and editing. DECKBench is built on a curated dataset of paper to slide pairs augmented with realistic, simulated editing instructions. Our evaluation protocol systematically assesses slide-level and deck-level fidelity, coherence, layout quality, and multi-turn instruction following. We further implement a modular multi-agent baseline system that decomposes the slide generation and editing task into paper parsing and summarization, slide planning, HTML creation, and iterative editing. Experimental results demonstrate that the proposed benchmark highlights strengths, exposes failure modes, and provides actionable insights for improving multi-agent slide generation and editing systems. Overall, this work establishes a standardized foundation for reproducible and comparable evaluation of academic presentation generation and editing. Code and data are publicly available at https://github.com/morgan-heisler/DeckBench . --- ### Semantic Waveforms for AI-Native 6G Networks Authors: Nour Hello, Mohamed Amine Hamoura, Francois Rivet, Emilio Calvanese Strinati Published: 2026-02-10 Categories: cs.NI, cs.AI, cs.ET arXiv: https://arxiv.org/abs/2602.13316v1 In this paper, we propose a semantic-aware waveform design framework for AI-native 6G networks that jointly optimizes physical layer resource usage and semantic communication efficiency and robustness, while explicitly accounting for the hardware constraints of RF chains. Our approach, called Orthogonal Semantic Sequency Division Multiplexing (OSSDM), introduces a parametrizable, orthogonal-base waveform design that enables controlled degradation of the wireless transmitted signal to preserve semantically significant content while minimizing resource consumption. We demonstrate that OSSDM not only reinforces semantic robustness against channel impairments but also improves semantic spectral efficiency by encoding meaningful information directly at the waveform level. Extensive numerical evaluations show that OSSDM outperforms conventional OFDM waveforms in spectral efficiency and semantic fidelity. The proposed semantic waveform co-design opens new research frontiers for AI-native, intelligent communication systems by enabling meaning-aware physical signal construction through the direct encoding of semantics at the waveform level. --- ### IDPruner: Harmonizing Importance and Diversity in Visual Token Pruning for MLLMs Authors: Yifan Tan, Yifu Sun, Shirui Huang, Hong Liu, Guanghua Yu, Jianchen Zhu, Yangdong Deng Published: 2026-02-10 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.13315v1 Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities, yet they encounter significant computational bottlenecks due to the massive volume of visual tokens. Consequently, visual token pruning, which substantially reduces the token count, has emerged as a critical technique for accelerating MLLM inference. Existing approaches focus on token importance, diversity, or an intuitive combination of both, without a principled framework for their optimal integration. To address this issue, we first conduct a systematic analysis to characterize the trade-off between token importance and semantic diversity. Guided by this analysis, we propose the \textbf{I}mportance and \textbf{D}iversity Pruner (\textbf{IDPruner}), which leverages the Maximal Marginal Relevance (MMR) algorithm to achieve a Pareto-optimal balance between these two objectives. Crucially, our method operates without requiring attention maps, ensuring full compatibility with FlashAttention and efficient deployment via one-shot pruning. We conduct extensive experiments across various model architectures and multimodal benchmarks, demonstrating that IDPruner achieves state-of-the-art performance and superior generalization across diverse architectures and tasks. Notably, on Qwen2.5-VL-7B-Instruct, IDPruner retains 95.18\% of baseline performance when pruning 75\% of the tokens, and still maintains 86.40\% even under an extreme 90\% pruning ratio. Our code is available at https://github.com/Tencent/AngelSlim. --- ### Sim2Radar: Toward Bridging the Radar Sim-to-Real Gap with VLM-Guided Scene Reconstruction Authors: Emily Bejerano, Federico Tondolo, Ayaan Qayyum, Xiaofan Yu, Xiaofan Jiang Published: 2026-02-10 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.13314v3 Millimeter-wave (mmWave) radar provides reliable perception in visually degraded indoor environments (e.g., smoke, dust, and low light), but learning-based radar perception is bottlenecked by the scarcity and cost of collecting and annotating large-scale radar datasets. We present Sim2Radar, an end-to-end framework that synthesizes training radar data directly from single-view RGB images, enabling scalable data generation without manual scene modeling. Sim2Radar reconstructs a material-aware 3D scene by combining monocular depth estimation, segmentation, and vision-language reasoning to infer object materials, then simulates mmWave propagation with a configurable physics-based ray tracer using Fresnel reflection models parameterized by ITU-R electromagnetic properties. Evaluated on real-world indoor scenes, Sim2Radar improves downstream 3D radar perception via transfer learning: pre-training a radar point-cloud object detection model on synthetic data and fine-tuning on real radar yields up to +3.7 3D AP (IoU 0.3), with gains driven primarily by improved spatial localization. These results suggest that physics-based, vision-driven radar simulation can provide effective geometric priors for radar learning and measurably improve performance under limited real-data supervision. --- ### Agentic Spatio-Temporal Grounding via Collaborative Reasoning Authors: Heng Zhao, Yew-Soon Ong, Joey Tianyi Zhou Published: 2026-02-10 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.13313v1 Spatio-Temporal Video Grounding (STVG) aims to retrieve the spatio-temporal tube of a target object or person in a video given a text query. Most existing approaches perform frame-wise spatial localization within a predicted temporal span, resulting in redundant computation, heavy supervision requirements, and limited generalization. Weakly-supervised variants mitigate annotation costs but remain constrained by the dataset-level train-and-fit paradigm with an inferior performance. To address these challenges, we propose the Agentic Spatio-Temporal Grounder (ASTG) framework for the task of STVG towards an open-world and training-free scenario. Specifically, two specialized agents SRA (Spatial Reasoning Agent) and TRA (Temporal Reasoning Agent) constructed leveraging on modern Multimoal Large Language Models (MLLMs) work collaboratively to retrieve the target tube in an autonomous and self-guided manner. Following a propose-and-evaluation paradigm, ASTG duly decouples spatio-temporal reasoning and automates the tube extraction, verification and temporal localization processes. With a dedicate visual memory and dialogue context, the retrieval efficiency is significantly enhanced. Experiments on popular benchmarks demonstrate the superiority of the proposed approach where it outperforms existing weakly-supervised and zero-shot approaches by a margin and is comparable to some of the fully-supervised methods. --- ### PeroMAS: A Multi-agent System of Perovskite Material Discovery Authors: Yishu Wang, Wei Liu, Yifan Li, Shengxiang Xu, Xujie Yuan, Ran Li, Yuyu Luo, Jia Zhu, Shimin Di, Min-Ling Zhang, Guixiang Li Published: 2026-02-10 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2602.13312v2 As a pioneer of the third-generation photovoltaic revolution, Perovskite Solar Cells (PSCs) are renowned for their superior optoelectronic performance and cost potential. The development process of PSCs is precise and complex, involving a series of closed-loop workflows such as literature retrieval, data integration, experimental design, and synthesis. However, existing AI perovskite approaches focus predominantly on discrete models, including material design, process optimization,and property prediction. These models fail to propagate physical constraints across the workflow, hindering end-to-end optimization. In this paper, we propose a multi-agent system for perovskite material discovery, named PeroMAS. We first encapsulated a series of perovskite-specific tools into Model Context Protocols (MCPs). By planning and invoking these tools, PeroMAS can design perovskite materials under multi-objective constraints, covering the entire process from literature retrieval and data extraction to property prediction and mechanism analysis. Furthermore, we construct an evaluation benchmark by perovskite human experts to assess this multi-agent system. Results demonstrate that, compared to single Large Language Model (LLM) or traditional search strategies, our system significantly enhances discovery efficiency. It successfully identified candidate materials satisfying multi-objective constraints. Notably, we verify PeroMAS's effectiveness in the physical world through real synthesis experiments. --- ### PeroMAS: A Multi-agent System of Perovskite Material Discovery Authors: Yishu Wang, Wei Liu, Yifan Li, Shengxiang Xu, Xujie Yuan, Ran Li, Yuyu Luo, Jia Zhu, Shimin Di, Min-Ling Zhang, Guixiang Li Published: 2026-02-10 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2602.13312v1 As a pioneer of the third-generation photovoltaic revolution, Perovskite Solar Cells (PSCs) are renowned for their superior optoelectronic performance and cost potential. The development process of PSCs is precise and complex, involving a series of closed-loop workflows such as literature retrieval, data integration, experimental design, and synthesis. However, existing AI perovskite approaches focus predominantly on discrete models, including material design, process optimization,and property prediction. These models fail to propagate physical constraints across the workflow, hindering end-to-end optimization. In this paper, we propose a multi-agent system for perovskite material discovery, named PeroMAS. We first encapsulated a series of perovskite-specific tools into Model Context Protocols (MCPs). By planning and invoking these tools, PeroMAS can design perovskite materials under multi-objective constraints, covering the entire process from literature retrieval and data extraction to property prediction and mechanism analysis. Furthermore, we construct an evaluation benchmark by perovskite human experts to assess this multi-agent system. Results demonstrate that, compared to single Large Language Model (LLM) or traditional search strategies, our system significantly enhances discovery efficiency. It successfully identified candidate materials satisfying multi-objective constraints. Notably, we verify PeroMAS's effectiveness in the physical world through real synthesis experiments. --- ### Visual Para-Thinker: Divide-and-Conquer Reasoning for Visual Comprehension Authors: Haoran Xu, Hongyu Wang, Jiaze Li, Shunpeng Chen, Zizhao Tong, Jianzhong Ju, Zhenbo Luo, Jian Luan Published: 2026-02-10 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.13310v2 Existing LLM test-time scaling laws emphasize the emergence of self-reflective behaviors through extended reasoning length. Nevertheless, this vertical scaling strategy often encounters plateaus in exploration as the model becomes locked into specific thinking pattern. By shifting from depth to parallelism, parallel thinking mitigates the narrowing of exploration. However, the extension of this paradigm to visual domain remains an open research question. In this paper, we first examine the role of visual partitioning in parallelized reasoning and subsequently propose two distinct strategies. Based on the above, we introduce Visual Para-Thinker, representing the inaugural parallel reasoning framework for MLLMs. To maintain path independence and promote diversity in reasoning, our approach integrates Pa-Attention alongside LPRoPE. Leveraging the vLLM framework, we have developed a native multimodal implementation that facilitates high-efficiency parallel processing. Empirical results on benchmark datasets such as V*, CountBench, RefCOCO, and HallusionBench confirm that Visual Para-Thinker successfully extends the benefits of parallel reasoning to the visual domain. --- ### Adaptive Value Decomposition: Coordinating a Varying Number of Agents in Urban Systems Authors: Yexin Li, Jinjin Guo, Haoyu Zhang, Yuhan Zhao, Yiwen Sun, Zihao Jiao Published: 2026-02-10 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2602.13309v1 Multi-agent reinforcement learning (MARL) provides a promising paradigm for coordinating multi-agent systems (MAS). However, most existing methods rely on restrictive assumptions, such as a fixed number of agents and fully synchronous action execution. These assumptions are often violated in urban systems, where the number of active agents varies over time, and actions may have heterogeneous durations, resulting in a semi-MARL setting. Moreover, while sharing policy parameters among agents is commonly adopted to improve learning efficiency, it can lead to highly homogeneous actions when a subset of agents make decisions concurrently under similar observations, potentially degrading coordination quality. To address these challenges, we propose Adaptive Value Decomposition (AVD), a cooperative MARL framework that adapts to a dynamically changing agent population. AVD further incorporates a lightweight mechanism to mitigate action homogenization induced by shared policies, thereby encouraging behavioral diversity and maintaining effective cooperation among agents. In addition, we design a training-execution strategy tailored to the semi-MARL setting that accommodates asynchronous decision-making when some agents act at different times. Experiments on real-world bike-sharing redistribution tasks in two major cities, London and Washington, D.C., demonstrate that AVD outperforms state-of-the-art baselines, confirming its effectiveness and generalizability. --- ### Learning to Select Like Humans: Explainable Active Learning for Medical Imaging Authors: Ifrat Ikhtear Uddin, Longwei Wang, Xiao Qin, Yang Zhou, KC Santosh Published: 2026-02-10 Categories: eess.IV, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2602.13308v3 Medical image analysis requires substantial labeled data for model training, yet expert annotation is expensive and time-consuming. Active learning (AL) addresses this challenge by strategically selecting the most informative samples for the annotation purpose, but traditional methods solely rely on predictive uncertainty while ignoring whether models learn from clinically meaningful features a critical requirement for clinical deployment. We propose an explainability-guided active learning framework that integrates spatial attention alignment into a sample acquisition process. Our approach advocates for a dual-criterion selection strategy combining: (i) classification uncertainty to identify informative examples, and (ii) attention misalignment with radiologist-defined regions-of-interest (ROIs) to target samples where the model focuses on incorrect features. By measuring misalignment between Grad-CAM attention maps and expert annotations using Dice similarity, our acquisition function judiciously identifies samples that enhance both predictive performance and spatial interpretability. We evaluate the framework using three expert-annotated medical imaging datasets, namely, BraTS (MRI brain tumors), VinDr-CXR (chest X-rays), and SIIM-COVID-19 (chest X-rays). Using only 570 strategically selected samples, our explainability-guided approach consistently outperforms random sampling across all the datasets, achieving 77.22% accuracy on BraTS, 52.37% on VinDr-CXR, and 52.66% on SIIM-COVID. Grad-CAM visualizations confirm that the models trained by our dual-criterion selection focus on diagnostically relevant regions, demonstrating that incorporating explanation guidance into sample acquisition yields superior data efficiency while maintaining clinical interpretability. --- ### Adaptive Physics Transformer with Fused Global-Local Attention for Subsurface Energy Systems Authors: Xin Ju, Nok Hei, Fung, Yuyan Zhang, Carl Jacquemyn, Matthew Jackson, Randolph Settgast, Sally M. Benson, Gege Wen Published: 2026-02-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.11208v2 The Earth's subsurface is a cornerstone of modern society, providing essential energy resources like hydrocarbons, geothermal, and minerals while serving as the primary reservoir for $CO_2$ sequestration. However, full physics numerical simulations of these systems are notoriously computationally expensive due to geological heterogeneity, high resolution requirements, and the tight coupling of physical processes with distinct propagation time scales. Here we propose the $\textbf{Adaptive Physics Transformer}$ (APT), a geometry-, mesh-, and physics-agnostic neural operator that explicitly addresses these challenges. APT fuses a graph-based encoder to extract high-resolution local heterogeneous features with a global attention mechanism to resolve long-range physical impacts. Our results demonstrate that APT outperforms state-of-the-art architectures in subsurface tasks across both regular and irregular grids with robust super-resolution capabilities. Notably, APT is the first architecture that learns directly from HR-adaptive mesh refinement simulations. We also demonstrate APT's favorable scaling behavior and cross-dataset learning capability, positioning it as a robust and scalable backbone for large-scale subsurface foundation model development. --- ### UltraLIF: Fully Differentiable Spiking Neural Networks via Ultradiscretization and Max-Plus Algebra Authors: Jose Marie Antonio Miñoza Published: 2026-02-10 Categories: cs.LG, cs.AI, cs.CV, math.RA, q-bio.NC arXiv: https://arxiv.org/abs/2602.11206v1 Spiking Neural Networks (SNNs) offer energy-efficient, biologically plausible computation but suffer from non-differentiable spike generation, necessitating reliance on heuristic surrogate gradients. This paper introduces UltraLIF, a principled framework that replaces surrogate gradients with ultradiscretization, a mathematical formalism from tropical geometry providing continuous relaxations of discrete dynamics. The central insight is that the max-plus semiring underlying ultradiscretization naturally models neural threshold dynamics: the log-sum-exp function serves as a differentiable soft-maximum that converges to hard thresholding as a learnable temperature parameter $\eps \to 0$. Two neuron models are derived from distinct dynamical systems: UltraLIF from the LIF ordinary differential equation (temporal dynamics) and UltraDLIF from the diffusion equation modeling gap junction coupling across neuronal populations (spatial dynamics). Both yield fully differentiable SNNs trainable via standard backpropagation with no forward-backward mismatch. Theoretical analysis establishes pointwise convergence to classical LIF dynamics with quantitative error bounds and bounded non-vanishing gradients. Experiments on six benchmarks spanning static images, neuromorphic vision, and audio demonstrate improvements over surrogate gradient baselines, with gains most pronounced in single-timestep ($T{=}1$) settings on neuromorphic and temporal datasets. An optional sparsity penalty enables significant energy reduction while maintaining competitive accuracy. --- ### Zero-Sacrifice Persistent-Robustness Adversarial Defense for Pre-Trained Encoders Authors: Zhuxin Lei, Ziyuan Yang, Yi Zhang Published: 2026-02-10 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2602.11204v1 The widespread use of publicly available pre-trained encoders from self-supervised learning (SSL) has exposed a critical vulnerability: their susceptibility to downstream-agnostic adversarial examples (DAEs), which are crafted without knowledge of the downstream tasks but capable of misleading downstream models. While several defense methods have been explored recently, they rely primarily on task-specific adversarial fine-tuning, which inevitably limits generalizability and causes catastrophic forgetting and deteriorates benign performance. Different with previous works, we propose a more rigorous defense goal that requires only a single tuning for diverse downstream tasks to defend against DAEs and preserve benign performance. To achieve this defense goal, we introduce Zero-Sacrifice Persistent-Robustness Adversarial Defense (ZePAD), which is inspired by the inherent sensitivity of neural networks to data characteristics. Specifically, ZePAD is a dual-branch structure, which consists of a Multi-Pattern Adversarial Enhancement Branch (MPAE-Branch) that uses two adversarially fine-tuned encoders to strengthen adversarial resistance. The Benign Memory Preservation Branch (BMP-Branch) is trained on local data to ensure adversarial robustness does not compromise benign performance. Surprisingly, we find that ZePAD can directly detect DAEs by evaluating branch confidence, without introducing any adversarial exsample identification task during training. Notably, by enriching feature diversity, our method enables a single adversarial fine-tuning to defend against DAEs across downstream tasks, thereby achieving persistent robustness. Extensive experiments on 11 SSL methods and 6 datasets validate its effectiveness. In certain cases, it achieves a 29.20% improvement in benign performance and a 73.86% gain in adversarial robustness, highlighting its zero-sacrifice property. --- ### Flash-SD-KDE: Accelerating SD-KDE with Tensor Cores Authors: Elliot L. Epstein, Rajat Vadiraj Dwaraknath, John Winnicki Published: 2026-02-10 Categories: cs.DC, cs.LG arXiv: https://arxiv.org/abs/2602.10378v1 Score-debiased kernel density estimation (SD-KDE) achieves improved asymptotic convergence rates over classical KDE, but its use of an empirical score has made it significantly slower in practice. We show that by re-ordering the SD-KDE computation to expose matrix-multiplication structure, Tensor Cores can be used to accelerate the GPU implementation. On a 32k-sample 16-dimensional problem, our approach runs up to $47\times$ faster than a strong SD-KDE GPU baseline and $3{,}300\times$ faster than scikit-learn's KDE. On a larger 1M-sample 16-dimensional task evaluated on 131k queries, Flash-SD-KDE completes in $2.3$ s on a single GPU, making score-debiased density estimation practical at previously infeasible scales. --- ### Hardware Co-Design Scaling Laws via Roofline Modelling for On-Device LLMs Authors: Luoyang Sun, Jiwen Jiang, Yifeng Ding, Fengfa Li, Yan Song, Haifeng Zhang, Jian Ying, Lei Ren, Kun Zhan, Wei Chen, Yan Xie, Cheng Deng Published: 2026-02-10 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2602.10377v1 Vision-Language-Action Models (VLAs) have emerged as a key paradigm of Physical AI and are increasingly deployed in autonomous vehicles, robots, and smart spaces. In these resource-constrained on-device settings, selecting an appropriate large language model (LLM) backbone is a critical challenge: models must balance accuracy with strict inference latency and hardware efficiency constraints. This makes hardware-software co-design a game-changing requirement for on-device LLM deployment, where each hardware platform demands a tailored architectural solution. We propose a hardware co-design law that jointly captures model accuracy and inference performance. Specifically, we model training loss as an explicit function of architectural hyperparameters and characterise inference latency via roofline modelling. We empirically evaluate 1,942 candidate architectures on NVIDIA Jetson Orin, training 170 selected models for 10B tokens each to fit a scaling law relating architecture to training loss. By coupling this scaling law with latency modelling, we establish a direct accuracy-latency correspondence and identify the Pareto frontier for hardware co-designed LLMs. We further formulate architecture search as a joint optimisation over precision and performance, deriving feasible design regions under industrial hardware and application budgets. Our approach reduces architecture selection from months to days. At the same latency as Qwen2.5-0.5B on the target hardware, our co-designed architecture achieves 19.42% lower perplexity on WikiText-2. To our knowledge, this is the first principled and operational framework for hardware co-design scaling laws in on-device LLM deployment. We will make the code and related checkpoints publicly available. --- ### Simple LLM Baselines are Competitive for Model Diffing Authors: Elias Kempf, Simon Schrodi, Bartosz Cywiński, Thomas Brox, Neel Nanda, Arthur Conmy Published: 2026-02-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.10371v1 Standard LLM evaluations only test capabilities or dispositions that evaluators designed them for, missing unexpected differences such as behavioral shifts between model revisions or emergent misaligned tendencies. Model diffing addresses this limitation by automatically surfacing systematic behavioral differences. Recent approaches include LLM-based methods that generate natural language descriptions and sparse autoencoder (SAE)-based methods that identify interpretable features. However, no systematic comparison of these approaches exists nor are there established evaluation criteria. We address this gap by proposing evaluation metrics for key desiderata (generalization, interestingness, and abstraction level) and use these to compare existing methods. Our results show that an improved LLM-based baseline performs comparably to the SAE-based method while typically surfacing more abstract behavioral differences. --- ### Causal Effect Estimation with Learned Instrument Representations Authors: Frances Dean, Jenna Fields, Radhika Bhalerao, Marie Charpignon, Ahmed Alaa Published: 2026-02-10 Categories: stat.ML, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2602.10370v2 Instrumental variable (IV) methods mitigate bias from unobserved confounding in observational causal inference but rely on the availability of a valid instrument, which can often be difficult or infeasible to identify in practice. In this paper, we propose a representation learning approach that constructs instrumental representations from observed covariates, which enable IV-based estimation even in the absence of an explicit instrument. Our model (ZNet) achieves this through an architecture that mirrors the structural causal model of IVs; it decomposes the ambient feature space into confounding and instrumental components, and is trained by enforcing empirical moment conditions corresponding to the defining properties of valid instruments (i.e., relevance, exclusion restriction, and instrumental unconfoundedness). Importantly, ZNet is compatible with a wide range of downstream two-stage IV estimators of causal effects. Our experiments demonstrate that ZNet can (i) recover ground-truth instruments when they already exist in the ambient feature space and (ii) construct latent instruments in the embedding space when no explicit IVs are available. Our work suggests when ZNet can be used as a module for causal inference in general observational settings. --- ### LiveMedBench: A Contamination-Free Medical Benchmark for LLMs with Automated Rubric Evaluation Authors: Zhiling Yan, Dingjie Song, Zhe Fang, Yisheng Ji, Xiang Li, Quanzheng Li, Lichao Sun Published: 2026-02-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2602.10367v1 The deployment of Large Language Models (LLMs) in high-stakes clinical settings demands rigorous and reliable evaluation. However, existing medical benchmarks remain static, suffering from two critical limitations: (1) data contamination, where test sets inadvertently leak into training corpora, leading to inflated performance estimates; and (2) temporal misalignment, failing to capture the rapid evolution of medical knowledge. Furthermore, current evaluation metrics for open-ended clinical reasoning often rely on either shallow lexical overlap (e.g., ROUGE) or subjective LLM-as-a-Judge scoring, both inadequate for verifying clinical correctness. To bridge these gaps, we introduce LiveMedBench, a continuously updated, contamination-free, and rubric-based benchmark that weekly harvests real-world clinical cases from online medical communities, ensuring strict temporal separation from model training data. We propose a Multi-Agent Clinical Curation Framework that filters raw data noise and validates clinical integrity against evidence-based medical principles. For evaluation, we develop an Automated Rubric-based Evaluation Framework that decomposes physician responses into granular, case-specific criteria, achieving substantially stronger alignment with expert physicians than LLM-as-a-Judge. To date, LiveMedBench comprises 2,756 real-world cases spanning 38 medical specialties and multiple languages, paired with 16,702 unique evaluation criteria. Extensive evaluation of 38 LLMs reveals that even the best-performing model achieves only 39.2%, and 84% of models exhibit performance degradation on post-cutoff cases, confirming pervasive data contamination risks. Error analysis further identifies contextual application-not factual knowledge-as the dominant bottleneck, with 35-48% of failures stemming from the inability to tailor medical knowledge to patient-specific constraints. --- ### Comp2Comp: Open-Source Software with FDA-Cleared Artificial Intelligence Algorithms for Computed Tomography Image Analysis Authors: Adrit Rao, Malte Jensen, Andrea T. Fisher, Louis Blankemeier, Pauline Berens, Arash Fereydooni, Seth Lirette, Eren Alkan, Felipe C. Kitamura, Juan M. Zambrano Chaves, Eduardo Reis, Arjun Desai et al. Published: 2026-02-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.10364v1 Artificial intelligence allows automatic extraction of imaging biomarkers from already-acquired radiologic images. This paradigm of opportunistic imaging adds value to medical imaging without additional imaging costs or patient radiation exposure. However, many open-source image analysis solutions lack rigorous validation while commercial solutions lack transparency, leading to unexpected failures when deployed. Here, we report development and validation for two of the first fully open-sourced, FDA-510(k)-cleared deep learning pipelines to mitigate both challenges: Abdominal Aortic Quantification (AAQ) and Bone Mineral Density (BMD) estimation are both offered within the Comp2Comp package for opportunistic analysis of computed tomography scans. AAQ segments the abdominal aorta to assess aneurysm size; BMD segments vertebral bodies to estimate trabecular bone density and osteoporosis risk. AAQ-derived maximal aortic diameters were compared against radiologist ground-truth measurements on 258 patient scans enriched for abdominal aortic aneurysms from four external institutions. BMD binary classifications (low vs. normal bone density) were compared against concurrent DXA scan ground truths obtained on 371 patient scans from four external institutions. AAQ had an overall mean absolute error of 1.57 mm (95% CI 1.38-1.80 mm). BMD had a sensitivity of 81.0% (95% CI 74.0-86.8%) and specificity of 78.4% (95% CI 72.3-83.7%). Comp2Comp AAQ and BMD demonstrated sufficient accuracy for clinical use. Open-sourcing these algorithms improves transparency of typically opaque FDA clearance processes, allows hospitals to test the algorithms before cumbersome clinical pilots, and provides researchers with best-in-class methods. --- ### ENIGMA: EEG-to-Image in 15 Minutes Using Less Than 1% of the Parameters Authors: Reese Kneeland, Wangshu Jiang, Ugo Bruzadin Nunes, Paul Steven Scotti, Arnaud Delorme, Jonathan Xu Published: 2026-02-10 Categories: q-bio.NC, cs.AI, cs.CV, cs.HC arXiv: https://arxiv.org/abs/2602.10361v1 To be practical for real-life applications, models for brain-computer interfaces must be easily and quickly deployable on new subjects, effective on affordable scanning hardware, and small enough to run locally on accessible computing resources. To directly address these current limitations, we introduce ENIGMA, a multi-subject electroencephalography (EEG)-to-Image decoding model that reconstructs seen images from EEG recordings and achieves state-of-the-art (SOTA) performance on the research-grade THINGS-EEG2 and consumer-grade AllJoined-1.6M benchmarks, while fine-tuning effectively on new subjects with as little as 15 minutes of data. ENIGMA boasts a simpler architecture and requires less than 1% of the trainable parameters necessary for previous approaches. Our approach integrates a subject-unified spatio-temporal backbone along with a set of multi-subject latent alignment layers and an MLP projector to map raw EEG signals to a rich visual latent space. We evaluate our approach using a broad suite of image reconstruction metrics that have been standardized in the adjacent field of fMRI-to-Image research, and we describe the first EEG-to-Image study to conduct extensive behavioral evaluations of our reconstructions using human raters. Our simple and robust architecture provides a significant performance boost across both research-grade and consumer-grade EEG hardware, and a substantial improvement in fine-tuning efficiency and inference cost. Finally, we provide extensive ablations to determine the architectural choices most responsible for our performance gains in both single and multi-subject cases across multiple benchmark datasets. Collectively, our work provides a substantial step towards the development of practical brain-computer interface applications. --- ### Beyond Calibration: Confounding Pathology Limits Foundation Model Specificity in Abdominal Trauma CT Authors: Jineel H Raythatha, Shuchang Ye, Jeremy Hsu, Jinman Kim Published: 2026-02-10 Categories: eess.IV, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2602.10359v2 Purpose: Translating foundation models into clinical practice requires evaluating their performance under compound distribution shift, where severe class imbalance coexists with heterogeneous imaging appearances. This challenge is relevant for traumatic bowel injury, a rare but high-mortality diagnosis. We investigated whether specificity deficits in foundation models are associated with heterogeneity in the negative class. Methods: This retrospective study used the multi-institutional, RSNA Abdominal Traumatic Injury CT dataset (2019-2023), comprising scans from 23 centres. Two foundation models (MedCLIP, zero-shot; RadDINO, linear probe) were compared against three task-specific approaches (CNN, Transformer, Ensemble). Models were trained on 3,147 patients (2.3% bowel injury prevalence) and evaluated on an enriched 100-patient test set. To isolate negative-class effects, specificity was assessed in patients without bowel injury who had concurrent solid organ injury (n=58) versus no abdominal pathology (n=50). Results: Foundation models achieved equivalent discrimination to task-specific models (AUC, 0.64-0.68 versus 0.58-0.64) with higher sensitivity (79-91% vs 41-74%) but lower specificity (33-50% vs 50-88%). All models demonstrated high specificity in patients without abdominal pathology (84-100%). When solid organ injuries were present, specificity declined substantially for foundation models (50-51 percentage points) compared with smaller reductions of 12-41 percentage points for task-specific models. Conclusion: Foundation models matched task-specific discrimination without task-specific training, but their specificity deficits were driven primarily by confounding negative-class heterogeneity rather than prevalence alone. Susceptibility to negative-class heterogeneity decreased progressively with labelled training, suggesting adaptation is required before clinical implementation. --- ### Theoretical Analysis of Contrastive Learning under Imbalanced Data: From Training Dynamics to a Pruning Solution Authors: Haixu Liao, Yating Zhou, Songyang Zhang, Meng Wang, Shuai Zhang Published: 2026-02-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.10357v1 Contrastive learning has emerged as a powerful framework for learning generalizable representations, yet its theoretical understanding remains limited, particularly under imbalanced data distributions that are prevalent in real-world applications. Such an imbalance can degrade representation quality and induce biased model behavior, yet a rigorous characterization of these effects is lacking. In this work, we develop a theoretical framework to analyze the training dynamics of contrastive learning with Transformer-based encoders under imbalanced data. Our results reveal that neuron weights evolve through three distinct stages of training, with different dynamics for majority features, minority features, and noise. We further show that minority features reduce representational capacity, increase the need for more complex architectures, and hinder the separation of ground-truth features from noise. Inspired by these neuron-level behaviors, we show that pruning restores performance degraded by imbalance and enhances feature separation, offering both conceptual insights and practical guidance. Major theoretical findings are validated through numerical experiments. --- ### Autonomous Continual Learning for Environment Adaptation of Computer-Use Agents Authors: Tianci Xue, Zeyi Liao, Tianneng Shi, Zilu Wang, Kai Zhang, Dawn Song, Yu Su, Huan Sun Published: 2026-02-10 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.10356v2 Real-world digital environments are highly diverse and dynamic. These characteristics cause agents to frequently encounter unseen environments and distribution shifts, making continual learning in such environments essential for computer-use agents (CUAs). However, a key challenge lies in obtaining high-quality and environment-grounded training data without relying on costly human annotation. In this work, we introduce ACuRL, an Autonomous Curriculum Reinforcement Learning framework that continually adapts agents to specific environments with zero human data. The agent first explores an environment to acquire initial experiences. During subsequent iterative training, a curriculum task generator leverages these experiences together with feedback from the previous iteration to synthesize new tasks tailored for the agent's current capabilities. To provide reliable reward signals, we introduce CUAJudge, a robust automatic evaluator for CUAs that achieves 93% agreement with human judgments. Empirically, our method effectively enables both intra-environment and cross-environment continual learning, yielding 3-29% absolute performance gains on the target environments without catastrophic forgetting on others. We also show that it can mitigate performance degradation under environment changes (e.g., version updates, platform migration, and resolution shifts). Further analyses show highly sparse updates (e.g., only 20% parameters), which helps explain the effective and robust adaptation. --- ### Physically Interpretable AlphaEarth Foundation Model Embeddings Enable LLM-Based Land Surface Intelligence Authors: Mashrekur Rahman Published: 2026-02-10 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.10354v1 Satellite foundation models produce dense embeddings whose physical interpretability remains poorly understood, limiting their integration into environmental decision systems. Using 12.1 million samples across the Continental United States (2017--2023), we first present a comprehensive interpretability analysis of Google AlphaEarth's 64-dimensional embeddings against 26 environmental variables spanning climate, vegetation, hydrology, temperature, and terrain. Combining linear, nonlinear, and attention-based methods, we show that individual embedding dimensions map onto specific land surface properties, while the full embedding space reconstructs most environmental variables with high fidelity (12 of 26 variables exceed $R^2 > 0.90$; temperature and elevation approach $R^2 = 0.97$). The strongest dimension-variable relationships converge across all three analytical methods and remain robust under spatial block cross-validation (mean $ΔR^2 = 0.017$) and temporally stable across all seven study years (mean inter-year correlation $r = 0.963$). Building on these validated interpretations, we then developed a Land Surface Intelligence system that implements retrieval-augmented generation over a FAISS-indexed embedding database of 12.1 million vectors, translating natural language environmental queries into satellite-grounded assessments. An LLM-as-Judge evaluation across 360 query--response cycles, using four LLMs in rotating generator, system, and judge roles, achieved weighted scores of $μ= 3.74 \pm 0.77$ (scale 1--5), with grounding ($μ= 3.93$) and coherence ($μ= 4.25$) as the strongest criteria. Our results demonstrate that satellite foundation model embeddings are physically structured representations that can be operationalized for environmental and geospatial intelligence. --- ### Learning Self-Interpretation from Interpretability Artifacts: Training Lightweight Adapters on Vector-Label Pairs Authors: Keenan Pepper, Alex McKenzie, Florin Pop, Stijn Servaes, Martin Leitgab, Mike Vaiana, Judd Rosenblatt, Michael S. A. Graziano, Diogo de Lucena Published: 2026-02-10 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.10352v2 Self-interpretation methods prompt language models to describe their own internal states, but remain unreliable due to hyperparameter sensitivity. We show that training lightweight adapters on interpretability artifacts, while keeping the LM entirely frozen, yields reliable self-interpretation across tasks and model families. A scalar affine adapter with just $d_\text{model}+1$ parameters suffices: trained adapters generate sparse autoencoder feature labels that outperform the training labels themselves (70% vs 50% generation scoring at 70B scale), identify topics with 94% recall@1 versus 1% for untrained baselines, and decode bridge entities in multi-hop reasoning that appear in neither prompt nor response, surfacing implicit reasoning without chain-of-thought. The learned bias vector alone accounts for 85% of improvement, and simpler adapters generalize better than more expressive alternatives. Controlling for model knowledge via prompted descriptions, we find self-interpretation gains outpace capability gains from 7B to 72B parameters. Our results demonstrate that self-interpretation improves with scale, without modifying the model being interpreted. --- ### When Less Is More? Diagnosing ASR Predictions in Sardinian via Layer-Wise Decoding Authors: Domenico De Cristofaro, Alessandro Vietti, Marianne Pouplier, Aleese Block Published: 2026-02-10 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.10350v1 Recent studies have shown that intermediate layers in multilingual speech models often encode more phonetically accurate representations than the final output layer. In this work, we apply a layer-wise decoding strategy to a pretrained Wav2Vec2 model to investigate how phoneme-level predictions evolve across encoder layers, focusing on Campidanese Sardinian, a low-resource language. We show that truncating upper transformer layers leads to improved Phoneme Error Rates (PER), with the best performance achieved not at the final layer, but two layers earlier. Through fine-grained alignment analysis, we find that intermediate predictions better preserve segmental identity, avoid overgeneration, and reduce certain classes of phonological errors. We also introduce the notion of regressive errors, cases where correct predictions at intermediate layers are overwritten by errors at the final layer. These regressions highlight the limitations of surface-level error metrics and reveal how deeper layers may generalize or abstract away from acoustic detail. Our findings support the use of early-layer probing as a diagnostic tool for ASR models, particularly in low-resource settings where standard evaluation metrics may fail to capture linguistically meaningful behavior. --- ### Geometry-Aware Decoding with Wasserstein-Regularized Truncation and Mass Penalties for Large Language Models Authors: Arash Gholami Davoodi, Navid Rezazadeh, Seyed Pouyan Mousavi Davoudi, Pouya Pezeshkpour Published: 2026-02-10 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2602.10346v2 Large language models (LLMs) must balance diversity and creativity against logical coherence in open-ended generation. Existing truncation-based samplers are effective but largely heuristic, relying mainly on probability mass and entropy while ignoring semantic geometry of the token space. We present Top-W, a geometry-aware truncation rule that uses Wasserstein distance-defined over token-embedding geometry-to keep the cropped distribution close to the original, while explicitly balancing retained probability mass against the entropy of the kept set. Our theory yields a simple closed-form structure for the fixed-potential subset update: depending on the mass-entropy trade-off, the optimal crop either collapses to a single token or takes the form of a one-dimensional prefix that can be found efficiently with a linear scan. We implement Top-W using efficient geometry-based potentials (nearest-set or k-NN) and pair it with an alternating decoding routine that keeps the standard truncation-and-sampling interface unchanged. Extensive experiments on four benchmarks (GSM8K, GPQA, AlpacaEval, and MT-Bench) across three instruction-tuned models show that Top-W consistently outperforms prior state-of-the-art decoding approaches achieving up to 33.7% improvement. Moreover, we find that Top-W not only improves accuracy-focused performance, but also boosts creativity under judge-based open-ended evaluation. --- ### Identifying Evidence-Based Nudges in Biomedical Literature with Large Language Models Authors: Jaydeep Chauhan, Mark Seidman, Pezhman Raeisian Parvari, Zhi Zheng, Zina Ben-Miled, Cristina Barboi, Andrew Gonzalez, Malaz Boustani Published: 2026-02-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.10345v1 We present a scalable, AI-powered system that identifies and extracts evidence-based behavioral nudges from unstructured biomedical literature. Nudges are subtle, non-coercive interventions that influence behavior without limiting choice, showing strong impact on health outcomes like medication adherence. However, identifying these interventions from PubMed's 8 million+ articles is a bottleneck. Our system uses a novel multi-stage pipeline: first, hybrid filtering (keywords, TF-IDF, cosine similarity, and a "nudge-term bonus") reduces the corpus to about 81,000 candidates. Second, we use OpenScholar (quantized LLaMA 3.1 8B) to classify papers and extract structured fields like nudge type and target behavior in a single pass, validated against a JSON schema. We evaluated four configurations on a labeled test set (N=197). The best setup (Title/Abstract/Intro) achieved a 67.0% F1 score and 72.0% recall, ideal for discovery. A high-precision variant using self-consistency (7 randomized passes) achieved 100% precision with 12% recall, demonstrating a tunable trade-off for high-trust use cases. This system is being integrated into Agile Nudge+, a real-world platform, to ground LLM-generated interventions in peer-reviewed evidence. This work demonstrates interpretable, domain-specific retrieval pipelines for evidence synthesis and personalized healthcare. --- ### Monte Carlo Maximum Likelihood Reconstruction for Digital Holography with Speckle Authors: Xi Chen, Arian Maleki, Shirin Jalali Published: 2026-02-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.10344v1 In coherent imaging, speckle is statistically modeled as multiplicative noise, posing a fundamental challenge for image reconstruction. While maximum likelihood estimation (MLE) provides a principled framework for speckle mitigation, its application to coherent imaging system such as digital holography with finite apertures is hindered by the prohibitive cost of high-dimensional matrix inversion, especially at high resolutions. This computational burden has prevented the use of MLE-based reconstruction with physically accurate aperture modeling. In this work, we propose a randomized linear algebra approach that enables scalable MLE optimization without explicit matrix inversions in gradient computation. By exploiting the structural properties of sensing matrix and using conjugate gradient for likelihood gradient evaluation, the proposed algorithm supports accurate aperture modeling without the simplifying assumptions commonly imposed for tractability. We term the resulting method projected gradient descent with Monte Carlo estimation (PGD-MC). The proposed PGD-MC framework (i) demonstrates robustness to diverse and physically accurate aperture models, (ii) achieves substantial improvements in reconstruction quality and computational efficiency, and (iii) scales effectively to high-resolution digital holography. Extensive experiments incorporating three representative denoisers as regularization show that PGD-MC provides a flexible and effective MLE-based reconstruction framework for digital holography with finite apertures, consistently outperforming prior Plug-and-Play model-based iterative reconstruction methods in both accuracy and speed. Our code is available at: https://github.com/Computational-Imaging-RU/MC_Maximum_Likelihood_Digital_Holography_Speckle. --- ### Conditional Uncertainty-Aware Political Deepfake Detection with Stochastic Convolutional Neural Networks Authors: Rafael-Petruţ Gardoş Published: 2026-02-10 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2602.10343v1 Recent advances in generative image models have enabled the creation of highly realistic political deepfakes, posing risks to information integrity, public trust, and democratic processes. While automated deepfake detectors are increasingly deployed in moderation and investigative pipelines, most existing systems provide only point predictions and fail to indicate when outputs are unreliable, being an operationally critical limitation in high-stakes political contexts. This work investigates conditional, uncertainty-aware political deepfake detection using stochastic convolutional neural networks within an empirical, decision-oriented reliability framework. Rather than treating uncertainty as a purely Bayesian construct, it is evaluated through observable criteria, including calibration quality, proper scoring rules, and its alignment with prediction errors under both global and confidence-conditioned analyses. A politically focused binary image dataset is constructed via deterministic metadata filtering from a large public real-synthetic corpus. Two pretrained CNN backbones (ResNet-18 and EfficientNet-B4) are fully fine-tuned for classification. Deterministic inference is compared with single-pass stochastic prediction, Monte Carlo dropout with multiple forward passes, temperature scaling, and ensemble-based uncertainty surrogates. Evaluation reports ROC-AUC, thresholded confusion matrices, calibration metrics, and generator-disjoint out-of-distribution performance. Results demonstrate that calibrated probabilistic outputs and uncertainty estimates enable risk-aware moderation policies. A systematic confidence-band analysis further clarifies when uncertainty provides operational value beyond predicted confidence, delineating both the benefits and limitations of uncertainty-aware deepfake detection in political settings. --- ### The Subjectivity of Respect in Police Traffic Stops: Modeling Community Perspectives in Body-Worn Camera Footage Authors: Preni Golazizian, Elnaz Rahmati, Jackson Trager, Zhivar Sourati, Nona Ghazizadeh, Georgios Chochlakis, Jose Alcocer, Kerby Bennett, Aarya Vijay Devnani, Parsa Hejabi, Harry G. Muttram, Akshay Kiran Padte et al. Published: 2026-02-10 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.10339v2 Traffic stops are among the most frequent police-civilian interactions, and body-worn cameras (BWCs) provide a unique record of how these encounters unfold. Respect is a central dimension of these interactions, shaping public trust and perceived legitimacy, yet its interpretation is inherently subjective and shaped by lived experience, rendering community-specific perspectives a critical consideration. Leveraging unprecedented access to Los Angeles Police Department BWC footage, we introduce the first large-scale traffic-stop dataset annotated with respect ratings and free-text rationales from multiple perspectives. By sampling annotators from police-affiliated, justice-system-impacted, and non-affiliated Los Angeles residents, we enable the systematic study of perceptual differences across diverse communities. To this end, we (i) develop a domain-specific evaluation rubric grounded in procedural justice theory, LAPD training materials, and extensive fieldwork; (ii) introduce a rubric-driven preference data construction framework for perspective-consistent alignment; and (iii) propose a perspective-aware modeling framework that predicts personalized respect ratings and generates annotator-specific rationales for both officers and civilian drivers from traffic-stop transcripts. Across all three annotator groups, our approach improves both rating prediction performance and rationale alignment. Our perspective-aware framework enables law enforcement to better understand diverse community expectations, providing a vital tool for building public trust and procedural legitimacy. --- ### Generalized Prediction-Powered Inference, with Application to Binary Classifier Evaluation Authors: Runjia Zou, Daniela Witten, Brian Williamson Published: 2026-02-10 Categories: stat.ME, stat.ML arXiv: https://arxiv.org/abs/2602.10332v1 In the partially-observed outcome setting, a recent set of proposals known as "prediction-powered inference" (PPI) involve (i) applying a pre-trained machine learning model to predict the response, and then (ii) using these predictions to obtain an estimator of the parameter of interest with asymptotic variance no greater than that which would be obtained using only the labeled observations. While existing PPI proposals consider estimators arising from M-estimation, in this paper we generalize PPI to any regular asymptotically linear estimator. Furthermore, by situating PPI within the context of an existing rich literature on missing data and semi-parametric efficiency theory, we show that while PPI does not achieve the semi-parametric efficiency lower bound outside of very restrictive and unrealistic scenarios, it can be viewed as a computationally-simple alternative to proposals in that literature. We exploit connections to that literature to propose modified PPI estimators that can handle three distinct forms of covariate distribution shift. Finally, we illustrate these developments by constructing PPI estimators of true positive rate, false positive rate, and area under the curve via numerical studies. --- ### Efficient reduction of stellar contamination and noise in planetary transmission spectra using neural networks Authors: David S. Duque-Castaño, Lauren Flor-Torres, Jorge I. Zuluaga Published: 2026-02-10 Categories: astro-ph.EP, astro-ph.IM, cs.LG arXiv: https://arxiv.org/abs/2602.10330v5 The characterization of exoplanetary atmospheres has been transformed by the James Webb Space Telescope (JWST), whose infrared sensitivity enables transmission spectroscopy at unprecedented precision. However, stellar heterogeneities (e.g., spots and faculae) remain a dominant source of contamination that can bias atmospheric retrievals if not properly corrected. We present a methodology for reducing stellar contamination and instrument-specific noise from exoplanet transmission spectra using neural networks, in particular the so-called Denoising AutoEncoders (DAEs). Our goals are to enable fast, accurate corrections that improve the reliability of atmospheric parameter retrievals and to promote the use of unsupervised algorithms for efficient data processing. We designed and trained DAE architectures using large synthetic datasets of terrestrial (TRAPPIST-1e analogues) and sub-Neptune (K2-18b analogues) planets. Atmospheric retrieval experiments were then performed on contaminated spectra in order to compare our deep-learning approach against standard correction methods in terms of accuracy and computational cost. Our autoencoders successfully reconstruct uncontaminated spectra, preserving essential molecular features even in low-S/N regimes. In retrieval tests, the denoising autoencoder pre-processing yields atmospheric parameter estimates broadly comparable to those obtained with simultaneous stellar-contamination fitting. Notably, our method maintains a much lower computational cost, approximately one order of magnitude smaller. These results demonstrate that DAEs outperform conventional correction methods in computational efficiency while maintaining high accuracy, paving the way for their integration into future atmospheric characterization pipelines for both rocky and sub-Neptune exoplanets. --- ### Efficient reduction of stellar contamination and noise in planetary transmission spectra using neural networks Authors: David S. Duque-Castaño, Lauren Flor-Torres, Jorge I. Zuluaga Published: 2026-02-10 Categories: astro-ph.EP, astro-ph.IM, cs.LG arXiv: https://arxiv.org/abs/2602.10330v3 Context: The characterization of exoplanetary atmospheres has been transformed by the James Webb Space Telescope (JWST), whose infrared sensitivity enables transmission spectroscopy at unprecedented precision. However, stellar heterogeneities (e.g., spots and faculae) remain a dominant source of contamination that can bias atmospheric retrievals if not properly corrected. Aims: We present a methodology for reducing stellar contamination and instrument-specific noise from exoplanet transmission spectra using neural networks, in particular the so-called Denoising AutoEncoders (DAE). Our goals are to enable fast, accurate corrections that improve the reliability of atmospheric parameter retrievals and to promote the use of unsupervised algorithms for efficient data processing. Methods: We designed and trained DAE architectures using large synthetic datasets of terrestrial (TRAPPIST-1e analogues) and sub-Neptune (K2-18b analogues) planets. Atmospheric retrieval experiments were then performed on contaminated spectra in order to compare our deep-learning approach against standard correction methods in terms of accuracy and computational cost. Results: Our autoencoders successfully reconstruct uncontaminated spectra, preserving essential molecular features even in low-S/N regimes. In retrieval tests, the denoising autoencoder pre-processing reduces bias in retrieved abundance parameters compared to uncorrected observations. Notably, our method matches the accuracy of simultaneous stellar-contamination fitting while maintaining a much lower computational cost, typically one order of magnitude smaller. Conclusions: These results demonstrate that DAEs outperform conventional correction methods in computational efficiency while maintaining high accuracy, paving the way for their integration into future atmospheric characterization pipelines for both rocky and giant exoplanets. --- ### Are More Tokens Rational? Inference-Time Scaling in Language Models as Adaptive Resource Rationality Authors: Zhimin Hu, Riya Roshan, Sashank Varma Published: 2026-02-10 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2602.10329v1 Human reasoning is shaped by resource rationality -- optimizing performance under constraints. Recently, inference-time scaling has emerged as a powerful paradigm to improve the reasoning performance of Large Language Models by expanding test-time computation. Specifically, instruction-tuned (IT) models explicitly generate long reasoning steps during inference, whereas Large Reasoning Models (LRMs) are trained by reinforcement learning to discover reasoning paths that maximize accuracy. However, it remains unclear whether resource-rationality can emerge from such scaling without explicit reward related to computational costs. We introduce a Variable Attribution Task in which models infer which variables determine outcomes given candidate variables, input-output trials, and predefined logical functions. By varying the number of candidate variables and trials, we systematically manipulate task complexity. Both models exhibit a transition from brute-force to analytic strategies as complexity increases. IT models degrade on XOR and XNOR functions, whereas LRMs remain robust. These findings suggest that models can adjust their reasoning behavior in response to task complexity, even without explicit cost-based reward. It provides compelling evidence that resource rationality is an emergent property of inference-time scaling itself. --- ### Flow Matching with Uncertainty Quantification and Guidance Authors: Juyeop Han, Lukas Lao Beyer, Sertac Karaman Published: 2026-02-10 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2602.10326v2 Despite the remarkable success of sampling-based generative models such as flow matching, they can still produce samples of inconsistent or degraded quality. To assess sample reliability and generate higher-quality outputs, we propose uncertainty-aware flow matching (UA-Flow), a lightweight extension of flow matching that predicts the velocity field together with heteroscedastic uncertainty. UA-Flow estimates per-sample uncertainty by propagating velocity uncertainty through the flow dynamics. These uncertainty estimates act as a reliability signal for individual samples, and we further use them to steer generation via uncertainty-aware classifier guidance and classifier-free guidance. Experiments on image generation show that UA-Flow produces uncertainty signals more highly correlated with sample fidelity than baseline methods, and that uncertainty-guided sampling further improves generation quality. --- ### Discovering Differences in Strategic Behavior Between Humans and LLMs Authors: Caroline Wang, Daniel Kasenberg, Kim Stachenfeld, Pablo Samuel Castro Published: 2026-02-10 Categories: cs.AI, cs.CL, cs.CY, cs.HC arXiv: https://arxiv.org/abs/2602.10324v2 As Large Language Models (LLMs) are increasingly deployed in social and strategic scenarios, it becomes critical to understand where and why their behavior diverges from that of humans. While behavioral game theory (BGT) provides a framework for analyzing behavior, existing models do not fully capture the idiosyncratic behavior of humans or black-box, non-human agents like LLMs. We employ AlphaEvolve, a cutting-edge program discovery tool, to directly discover interpretable models of human and LLM behavior from data, thereby enabling open-ended discovery of structural factors driving human and LLM behavior. Our analysis on iterated rock-paper-scissors reveals that frontier LLMs can be capable of deeper strategic behavior than humans. These results provide a foundation for understanding structural differences driving differences in human and LLM behavior in strategic interactions. --- ### A Low-Rank Defense Method for Adversarial Attack on Diffusion Models Authors: Jiaxuan Zhu, Siyu Huang Published: 2026-02-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.10319v1 Recently, adversarial attacks for diffusion models as well as their fine-tuning process have been developed rapidly. To prevent the abuse of these attack algorithms from affecting the practical application of diffusion models, it is critical to develop corresponding defensive strategies. In this work, we propose an efficient defensive strategy, named Low-Rank Defense (LoRD), to defend the adversarial attack on Latent Diffusion Models (LDMs). LoRD introduces the merging idea and a balance parameter, combined with the low-rank adaptation (LoRA) modules, to detect and defend the adversarial samples. Based on LoRD, we build up a defense pipeline that applies the learned LoRD modules to help diffusion models defend against attack algorithms. Our method ensures that the LDM fine-tuned on both adversarial and clean samples can still generate high-quality images. To demonstrate the effectiveness of our approach, we conduct extensive experiments on facial and landscape images, and our method shows significantly better defense performance compared to the baseline methods. --- ### Uncertainty-Aware Ordinal Deep Learning for cross-Dataset Diabetic Retinopathy Grading Authors: Ali El Bellaj, Aya Benradi, Salman El Youssoufi, Taha El Marzouki, Mohammed-Amine Cheddadi Published: 2026-02-10 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2602.10315v1 Diabetes mellitus is a chronic metabolic disorder characterized by persistent hyperglycemia due to insufficient insulin production or impaired insulin utilization. One of its most severe complications is diabetic retinopathy (DR), a progressive retinal disease caused by microvascular damage, leading to hemorrhages, exudates, and potential vision loss. Early and reliable detection of DR is therefore critical for preventing irreversible blindness. In this work, we propose an uncertainty-aware deep learning framework for automated DR severity grading that explicitly models the ordinal nature of disease progression. Our approach combines a convolutional backbone with lesion-query attention pooling and an evidential Dirichlet-based ordinal regression head, enabling both accurate severity prediction and principled estimation of predictive uncertainty. The model is trained using an ordinal evidential loss with annealed regularization to encourage calibrated confidence under domain shift. We evaluate the proposed method on a multi-domain training setup combining APTOS, Messidor-2, and a subset of EyePACS fundus datasets. Experimental results demonstrate strong cross-dataset generalization, achieving competitive classification accuracy and high quadratic weighted kappa on held-out test sets, while providing meaningful uncertainty estimates for low-confidence cases. These results suggest that ordinal evidential learning is a promising direction for robust and clinically reliable diabetic retinopathy grading. --- ### Stop Training for the Worst: Progressive Unmasking Accelerates Masked Diffusion Training Authors: Jaeyeon Kim, Jonathan Geuter, David Alvarez-Melis, Sham Kakade, Sitan Chen Published: 2026-02-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.10314v2 Masked Diffusion Models (MDMs) have emerged as a promising approach for generative modeling in discrete spaces. By generating sequences in any order and allowing for parallel decoding, they enable fast inference and strong performance on non-causal tasks. However, this flexibility comes with a training complexity trade-off: MDMs train on an exponentially large set of masking patterns, which is not only computationally expensive, but also creates a train--test mismatch between the random masks used in training and the highly structured masks induced by inference-time unmasking. In this work, we propose Progressive UnMAsking (PUMA), a simple modification of the forward masking process that aligns training-time and inference-time masking patterns, thereby focusing optimization on inference-aligned masks and speeding up training. Empirically, PUMA speeds up pretraining at the 125M scale by $\approx 2.5\times$ and offers complementary advantages on top of common recipes like autoregressive initialization. We open-source our codebase at https://github.com/JaeyeonKim01/PUMA. --- ### Training-free retrieval-augmented generation with reinforced reasoning for flood damage nowcasting Authors: Lipai Huang, Kai Yin, Chia-Fu Liu, Ali Mostafavi Published: 2026-02-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.10312v2 We propose R2RAG-Flood, a training-free retrieval-augmented generation framework for flood damage nowcasting with reinforced reasoning. The framework builds a reasoning-centric knowledge base from labeled tabular records, where each sample includes structured predictors, a compact text-mode summary, and a model-generated reasoning trajectory. During inference, the target prompt is augmented with geographically local neighbors and selected free-shots to support case-based reasoning without task-specific fine-tuning. A two-stage procedure first determines damage occurrence and then refines severity within a three-level Property Damage Extent (PDE) classification, followed by a conservative downgrade check for weakly supported over-severe outputs. In a Hurricane Harvey case study in Harris County, Texas, the supervised tabular baseline achieves 0.714 overall accuracy and 0.859 accuracy on the damaged classes (medium and high PDE). Across seven LLM backbones, R2RAG-Flood achieves 0.613--0.668 overall accuracy and 0.757--0.896 accuracy on the damaged classes while providing a structured rationale for each prediction. Under the severity-per-cost metric used in this study, lighter R2RAG-Flood variants are more cost-efficient than the supervised baseline and larger LLM backbones. These results demonstrate the feasibility of a reasoning-centric, training-free pipeline for flood damage nowcasting in a realistic case-study setting. --- ### Confounding Robust Continuous Control via Automatic Reward Shaping Authors: Mateo Juliani, Mingxuan Li, Elias Bareinboim Published: 2026-02-10 Categories: cs.LG, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2602.10305v1 Reward shaping has been applied widely to accelerate Reinforcement Learning (RL) agents' training. However, a principled way of designing effective reward shaping functions, especially for complex continuous control problems, remains largely under-explained. In this work, we propose to automatically learn a reward shaping function for continuous control problems from offline datasets, potentially contaminated by unobserved confounding variables. Specifically, our method builds upon the recently proposed causal Bellman equation to learn a tight upper bound on the optimal state values, which is then used as the potentials in the Potential-Based Reward Shaping (PBRS) framework. Our proposed reward shaping algorithm is tested with Soft-Actor-Critic (SAC) on multiple commonly used continuous control benchmarks and exhibits strong performance guarantees under unobserved confounders. More broadly, our work marks a solid first step towards confounding robust continuous control from a causal perspective. Code for training our reward shaping functions can be found at https://github.com/mateojuliani/confounding_robust_cont_control. ---