# Research Papers — 2026-03 · bundle page 202 of 243 > Full-text markdown of papers #10051–10100 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/202 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/201.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/203.md) ### Spectral Probing of Feature Upsamplers in 2D-to-3D Scene Reconstruction Authors: Ling Xiao, Yuliang Xiu, Yue Chen, Guoming Wang, Toshihiko Yamasaki Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.05787v1 A typical 2D-to-3D pipeline takes multi-view images as input, where a Vision Foundation Model (VFM) extracts features that are spatially upsampled to dense representations for 3D reconstruction. If dense features across views preserve geometric consistency, differentiable rendering can recover an accurate 3D representation, making the feature upsampler a critical component. Recent learnable upsampling methods mainly aim to enhance spatial details, such as sharper geometry or richer textures, yet their impact on 3D awareness remains underexplored. To address this gap, we introduce a spectral diagnostic framework with six complementary metrics that characterize amplitude redistribution, structural spectral alignment, and directional stability. Across classical interpolation and learnable upsampling methods on CLIP and DINO backbones, we observe three key findings. First, structural spectral consistency (SSC/CSC) is the strongest predictor of NVS quality, whereas High-Frequency Spectral Slope Drift (HFSS) often correlates negatively with reconstruction performance, indicating that emphasizing high-frequency details alone does not necessarily improve 3D reconstruction. Second, geometry and texture respond to different spectral properties: Angular Energy Consistency (ADC) correlates more strongly with geometry-related metrics, while SSC/CSC influence texture fidelity slightly more than geometric accuracy. Third, although learnable upsamplers often produce sharper spatial features, they rarely outperform classical interpolation in reconstruction quality, and their effectiveness depends on the reconstruction model. Overall, our results indicate that reconstruction quality is more closely related to preserving spectral structure than to enhancing spatial detail, highlighting spectral consistency as an important principle for designing upsampling strategies in 2D-to-3D pipelines. --- ### Proof-of-Guardrail in AI Agents and What (Not) to Trust from It Authors: Xisen Jin, Michael Duan, Qin Lin, Aaron Chan, Zhenglun Chen, Junyi Du, Xiang Ren Published: 2026-03-06 Categories: cs.CR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2603.05786v2 As AI agents become widely deployed as online services, users often rely on an agent developer's claim about how safety is enforced, which introduces a threat where safety measures are falsely advertised. To address the threat, we propose proof-of-guardrail, a system that enables developers to provide cryptographic proof that a response is generated after a specific open-source guardrail. To generate proof, the developer runs the agent and guardrail inside a Trusted Execution Environment (TEE), which produces a TEE-signed attestation of guardrail code execution verifiable by any user offline. We implement proof-of-guardrail for OpenClaw agents and evaluate latency overhead and deployment cost. Proof-of-guardrail ensures integrity of guardrail execution while keeping the developer's agent private, but we also highlight a risk of deception about safety, for example, when malicious developers actively jailbreak the guardrail. Code and demo video: https://github.com/SaharaLabsAI/Verifiable-ClawGuard --- ### Visual Words Meet BM25: Sparse Auto-Encoder Visual Word Scoring for Image Retrieval Authors: Donghoon Han, Eunhwan Park, Seunghyeon Seo Published: 2026-03-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.05781v1 Dense image retrieval is accurate but offers limited interpretability and attribution, and it can be compute-intensive at scale. We present \textbf{BM25-V}, which applies Okapi BM25 scoring to sparse visual-word activations from a Sparse Auto-Encoder (SAE) on Vision Transformer patch features. Across a large gallery, visual-word document frequencies are highly imbalanced and follow a Zipfian-like distribution, making BM25's inverse document frequency (IDF) weighting well suited for suppressing ubiquitous, low-information words and emphasizing rare, discriminative ones. BM25-V retrieves high-recall candidates via sparse inverted-index operations and serves as an efficient first-stage retriever for dense reranking. Across seven benchmarks, BM25-V achieves Recall@200 $\geq$ 0.993, enabling a two-stage pipeline that reranks only $K{=}200$ candidates per query and recovers near-dense accuracy within $0.2$\% on average. An SAE trained once on ImageNet-1K transfers zero-shot to seven fine-grained benchmarks without fine-tuning, and BM25-V retrieval decisions are attributable to specific visual words with quantified IDF contributions. --- ### Balancing Domestic and Global Perspectives: Evaluating Dual-Calibration and LLM-Generated Nudges for Diverse News Recommendation Authors: Ruixuan Sun, Matthew Zent, Minzhu Zhao, Thanmayee Boyapati, Xinyi Li, Joseph A. Konstan Published: 2026-03-06 Categories: cs.IR, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2603.05780v1 In this study, we applied the ``personalized diversity nudge framework'' with the goal of expanding user reading coverage in terms of news locality (i.e., domestic and world news). We designed a novel topic-locality dual calibration algorithmic nudge and a large language model-based news personalization presentation nudge, then launched a 5-week real-user study with 120 U.S. news readers on the news recommendation experiment platform POPROX. With user interaction logs and survey responses, we found that algorithmic nudges can successfully increase exposure and consumption diversity, while the impact of LLM-based presentation nudges varied. User-level topic interest is a strong predictor of user clicks, while highlighting the relevance of news articles to prior read articles outperforms generic topic-based and no personalization. We also demonstrate that longitudinal exposure to calibrated news may shift readers' reading habits to value a balanced news digest from both domestic and world articles. Our results provide direction for future work on nudging for diverse consumption in news recommendation systems. --- ### Tutor Move Taxonomy: A Theory-Aligned Framework for Analyzing Instructional Moves in Tutoring Authors: Zhuqian Zhou, Kirk Vanacore, Tamisha Thompson, Jennifer St John, Rene Kizilcec Published: 2026-03-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.05778v1 Understanding what makes tutoring effective requires methods for systematically analyzing tutors' instructional actions during learning interactions. This paper presents a tutor move taxonomy designed to support large-scale analysis of tutoring dialogue within the National Tutoring Observatory. The taxonomy provides a structured annotation framework for labeling tutors' instructional moves during one-on-one tutoring sessions. We developed the taxonomy through a hybrid deductive-inductive process. First, we synthesized research from cognitive science, the learning sciences, classroom discourse analysis, and intelligent tutoring systems to construct a preliminary framework of tutoring moves. We then refined the taxonomy through iterative coding of authentic tutoring transcripts conducted by expert annotators with extensive instructional and qualitative research experience. The resulting taxonomy organizes tutoring behaviors into four categories: tutoring support, learning support, social-emotional and motivational support, and logistical support. Learning support moves are further organized along a spectrum of student engagement, distinguishing between moves that elicit student reasoning and those that provide direct explanation or answers. By defining tutoring dialogue in terms of discrete instructional actions, the taxonomy enables scalable annotation using AI, computational modeling of tutoring strategies, and empirical analysis of how tutoring behaviors relate to learning outcomes. --- ### PVminerLLM: Structured Extraction of Patient Voice from Patient-Generated Text using Large Language Models Authors: Samah Fodeh, Linhai Ma, Ganesh Puthiaraju, Srivani Talakokkul, Afshan Khan, Ashley Hagaman, Sarah Lowe, Aimee Roundtree Published: 2026-03-06 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.05776v1 Motivation: Patient-generated text contains critical information about patients' lived experiences, social circumstances, and engagement in care, including factors that strongly influence adherence, care coordination, and health equity. However, these patient voice signals are rarely available in structured form, limiting their use in patient-centered outcomes research and clinical quality improvement. Reliable extraction of such information is therefore essential for understanding and addressing non-clinical drivers of health outcomes at scale. Results: We introduce PVminer, a benchmark for structured extraction of patient voice, and propose PVminerLLM, a supervised fine-tuned large language model tailored to this task. Across multiple datasets and model sizes, PVminerLLM substantially outperforms prompt-based baselines, achieving up to 83.82% F1 for Code prediction, 80.74% F1 for Sub-code prediction, and 87.03% F1 for evidence Span extraction. Notably, strong performance is achieved even with smaller models, demonstrating that reliable patient voice extraction is feasible without extreme model scale. These results enable scalable analysis of social and experiential signals embedded in patient-generated text. Availability and Implementation: Code, evaluation scripts, and trained LLMs will be released publicly. Annotated datasets will be made available upon request for research use. Keywords: Large Language Models, Supervised Fine-Tuning, Medical Annotation, Patient-Generated Text, Clinical NLP --- ### First-Order Softmax Weighted Switching Gradient Method for Distributed Stochastic Minimax Optimization with Stochastic Constraints Authors: Zhankun Luo, Antesh Upadhyay, Sang Bin Moon, Abolfazl Hashemi Published: 2026-03-06 Categories: cs.LG, cs.DC arXiv: https://arxiv.org/abs/2603.05774v2 This paper addresses the distributed stochastic minimax optimization problem subject to stochastic constraints. We propose a novel first-order Softmax-Weighted Switching Gradient method tailored for federated learning. Under full client participation, our algorithm achieves the standard $\tilde{\mathcal{O}}(ε^{-4})$ oracle complexity to satisfy a unified bound $ε$ for both the optimality gap and feasibility tolerance. We extend our theoretical analysis to the practical partial participation regime by quantifying client sampling noise through a stochastic superiority assumption. Furthermore, by relaxing standard boundedness assumptions on the objective functions, we establish a strictly tighter lower bound for the softmax hyperparameter. We provide a unified error decomposition and establish a sharp $\mathcal{O}(\log\frac{1}δ)$ high-probability convergence guarantee. Ultimately, our framework demonstrates that a single-loop primal-only switching mechanism provides a stable alternative for optimizing worst-case client performance, effectively bypassing the hyperparameter sensitivity and convergence oscillations often encountered in traditional primal-dual or penalty-based approaches. We verify the efficacy of our algorithm via experiment on the Neyman-Pearson (NP) classification, fair classification, and federated safe reinforcement learning tasks. --- ### Knowing without Acting: The Disentangled Geometry of Safety Mechanisms in Large Language Models Authors: Jinman Wu, Yi Xie, Shen Lin, Shiqian Zhao, Xiaofeng Chen Published: 2026-03-06 Categories: cs.CR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.05773v2 Safety alignment is often conceptualized as a monolithic process wherein harmfulness detection automatically triggers refusal. However, the persistence of jailbreak attacks suggests a fundamental mechanistic decoupling. We propose the \textbf{\underline{D}}isentangled \textbf{\underline{S}}afety \textbf{\underline{H}}ypothesis \textbf{(DSH)}, positing that safety computation operates on two distinct subspaces: a \textit{Recognition Axis} ($\mathbf{v}_H$, ``Knowing'') and an \textit{Execution Axis} ($\mathbf{v}_R$, ``Acting''). Our geometric analysis reveals a universal ``Reflex-to-Dissociation'' evolution, where these signals transition from antagonistic entanglement in early layers to structural independence in deep layers. To validate this, we introduce \textit{Double-Difference Extraction} and \textit{Adaptive Causal Steering}. Using our curated \textsc{AmbiguityBench}, we demonstrate a causal double dissociation, effectively creating a state of ``Knowing without Acting.'' Crucially, we leverage this disentanglement to propose the \textbf{Refusal Erasure Attack (REA)}, which achieves State-of-the-Art attack success rates by surgically lobotomizing the refusal mechanism. Furthermore, we uncover a critical architectural divergence, contrasting the \textit{Explicit Semantic Control} of Llama3.1 with the \textit{Latent Distributed Control} of Qwen2.5. The code and dataset are available at https://anonymous.4open.science/r/DSH. --- ### Depth Charge: Jailbreak Large Language Models from Deep Safety Attention Heads Authors: Jinman Wu, Yi Xie, Shiqian Zhao, Xiaofeng Chen Published: 2026-03-06 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2603.05772v2 Currently, open-sourced large language models (OSLLMs) have demonstrated remarkable generative performance. However, as their structure and weights are made public, they are exposed to jailbreak attacks even after alignment. Existing attacks operate primarily at shallow levels, such as the prompt or embedding level, and often fail to expose vulnerabilities rooted in deeper model components, which creates a false sense of security for successful defense. In this paper, we propose \textbf{\underline{S}}afety \textbf{\underline{A}}ttention \textbf{\underline{H}}ead \textbf{\underline{A}}ttack (\textbf{SAHA}), an attention-head-level jailbreak framework that explores the vulnerability in deeper but insufficiently aligned attention heads. SAHA contains two novel designs. Firstly, we reveal that deeper attention layers introduce more vulnerability against jailbreak attacks. Based on this finding, \textbf{SAHA} introduces \textit{Ablation-Impact Ranking} head selection strategy to effectively locate the most vital layer for unsafe output. Secondly, we introduce a boundary-aware perturbation method, \textit{i.e. Layer-Wise Perturbation}, to probe the generation of unsafe content with minimal perturbation to the attention. This constrained perturbation guarantees higher semantic relevance with the target intent while ensuring evasion. Extensive experiments show the superiority of our method: SAHA improves ASR by 14\% over SOTA baselines, revealing the vulnerability of the attack surface on the attention head. Our code is available at https://anonymous.4open.science/r/SAHA. --- ### Layer-wise Instance Binding for Regional and Occlusion Control in Text-to-Image Diffusion Transformers Authors: Ruidong Chen, Yancheng Bai, Xuanpu Zhang, Jianhao Zeng, Lanjun Wang, Dan Song, Lei Sun, Xiangxiang Chu, Anan Liu Published: 2026-03-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.05769v1 Region-instructed layout control in text-to-image generation is highly practical, yet existing methods suffer from limitations: (i) training-based approaches inherit data bias and often degrade image quality, and (ii) current techniques struggle with occlusion order, limiting real-world usability. To address these issues, we propose LayerBind. By modeling regional generation as distinct layers and binding them during the generation, our method enables precise regional and occlusion controllability. Our motivation stems from the observation that spatial layout and occlusion are established at a very early denoising stage, suggesting that rearranging the early latent structure is sufficient to modify the final output. Building on this, we structure the scheme into two phases: instance initialization and subsequent semantic nursing. (1) First, leveraging the contextual sharing mechanism in multimodal joint attention, Layer-wise Instance Initialization creates per-instance branches that attend to their own regions while anchoring to the shared background. At a designated early step, these branches are fused according to the layer order to form a unified latent with a pre-established layout. (2) Then, Layer-wise Semantic Nursing reinforces regional details and maintains the occlusion order via a layer-wise attention enhancement. Specifically, a sequential layered attention path operates alongside the standard global path, with updates composited under a layer-transparency scheduler. LayerBind is training-free and plug-and-play, serving as a regional and occlusion controller across Diffusion Transformers. Beyond generation, it natively supports editable workflows, allowing for flexible modifications like changing instances or rearranging visible orders. Both qualitative and quantitative results demonstrate LayerBind's effectiveness, highlighting its strong potential for creative applications. --- ### Bridging Domains through Subspace-Aware Model Merging Authors: Levy Chaves, Chao Zhou, Rebekka Burkholz, Eduardo Valle, Sandra Avila Published: 2026-03-06 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2603.05768v2 Model merging integrates multiple task-specific models into a single consolidated one. Recent research has made progress in improving merging performance for in-distribution or multi-task scenarios, but domain generalization in model merging remains underexplored. We investigate how merging models fine-tuned on distinct domains affects generalization to unseen domains. Through an analysis of parameter competition in the task matrix using singular value decomposition, we show that merging models trained under different distribution shifts induces stronger conflicts between their subspaces compared to traditional multi-task settings. To mitigate this issue, we propose SCORE (Subspace COnflict-Resolving mErging), a method designed to alleviate such singular subspace conflicts. SCORE finds a shared orthogonal basis by computing the principal components of the concatenated leading singular vectors of all models. It then projects each task matrix into the shared basis, pruning off-diagonal components to remove conflicting singular directions. SCORE consistently outperforms, on average, existing model merging approaches in domain generalization settings across a variety of architectures and model scales, demonstrating its effectiveness and scalability. --- ### BASIS: Balanced Activation Sketching with Invariant Scalars for "Ghost Backpropagation" Authors: Vladimer Khasia Published: 2026-03-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.16324v1 The activation memory required for exact backpropagation scales linearly with network depth, context length, and feature dimensionality, forming an O(L * BN ) spatial bottleneck (where B is the sequence-batch cardinality and N is the feature dimension). This constraint historically throttles the scaling of deep neural networks. While randomized automatic differentiation attempts to mitigate this, it historically suffers from catastrophic variance. In this paper, we introduce BASIS (Balanced Activation Sketching with Invariant Scalars), an efficient backpropagation algorithm that fully decouples activation memory from the batch and sequence dimensions. BASIS propagates the exact error signal (dX) to preserve flawless gradient flow, but computes the weight updates (dW) using massively compressed rank-R tensors. To solve the foundational instability of sketched gradients, we propose two novel mechanisms: Balanced Hashing, which strictly eliminates off-diagonal collision variance, and Invariant Scalars, a principled bias-variance tradeoff that deterministically preserves the exact continuous energy norm of the spatial geometry. Theoretically, BASIS reduces activation memory to O(L * RN ) and heavily decreases the backward pass matrix-multiplication footprint. Empirically, training a GPT architecture for 50,000 steps validates our theoretical guarantees: at R = 32, BASIS achieves parity with (and marginally outperforms) exact backpropagation validation loss (6.575 vs. 6.616), acting as an implicit regularizer. Remarkably, the stabilized magnitude trajectory allows the model to converge smoothly even under extreme spatial compression (R = 1), proving the extreme robustness of the estimator. The code is available at https://github.com/VladimerKhasia/basis --- ### Why Smaller Is Slower? Dimensional Misalignment in Compressed LLMs Authors: Jihao Xin, Tian Lyu, Qilong Pan, Kesen Wang, Marco Canini Published: 2026-03-05 Categories: cs.DC, cs.AI arXiv: https://arxiv.org/abs/2604.09595v1 Post-training compression reduces LLM parameter counts but often produces irregular tensor dimensions that degrade GPU performance -- a phenomenon we call \emph{dimensional misalignment}. We present a full-stack analysis tracing root causes at three levels: framework, library, and hardware. The key insight is that model inference becomes slower because the resulting dimensions are unfriendly with the GPU execution stack. For example, compressing Llama-3-8B with activation-aware singular value decomposition (ASVD) has 15\% fewer parameters yet runs no faster than the uncompressed baseline, because 95\% of its dimensions are misaligned. We propose \textbf{GAC} (GPU-Aligned Compression), a new compression paradigm that wraps any dimension-reducing compressor and re-selects hardware-aligned dimensions via multi-choice knapsack optimization under the same parameter budget. We evaluate GAC on Llama-3-8B with ASVD and LLM-Pruner, achieving 100\% alignment and recovering up to 1.5$\times$ speedup while preserving model quality. --- ### Spatial Competence Benchmark Authors: Jash Vira, Ashley Harris Published: 2026-03-05 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.09594v1 Spatial competence is the quality of maintaining a consistent internal representation of an environment and using it to infer discrete structure and plan actions under constraints. Prevailing spatial evaluations for large models are limited to probing isolated primitives through 3D transformations or visual question answering. We introduce the Spatial Competence Benchmark (SCBench), spanning three hierarchical capability buckets whose tasks require executable outputs verified by deterministic checkers or simulator-based evaluators. On SCBench, three frontier models exhibit monotonically decreasing accuracy up the capability ladder. Sweeping output-token caps shows that accuracy gains concentrate at low budgets and saturate quickly, and failures are dominated by locally plausible geometry that breaks global constraints. We release the task generators, verifiers, and visualisation tooling. --- ### ReVEL: Multi-Turn Reflective LLM-Guided Heuristic Evolution via Structured Performance Feedback Authors: Cuong Van Duc, Minh Nguyen Dinh Tuan, Tam Vu Duc, Tung Vu Duy, Son Nguyen Van, Hanh Nguyen Thi, Binh Huynh Thi Thanh Published: 2026-03-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.04940v3 Designing effective heuristics for NP-hard combinatorial optimization problems remains challenging and often requires substantial domain expertise. Recent LLM-guided evolutionary methods have shown promise for automated heuristic generation, but most existing approaches refine heuristics independently or through limited pairwise feedback. We propose ReVEL: Multi-Turn Reflective LLM-Guided Heuristic Evolution via Structured Performance Feedback, a framework for group-wise multi-turn heuristic refinement. ReVEL organizes heuristics into behavior-aware reflective groups, including similarity-driven groups for localized refinement and diversity-driven groups for exploratory search. Within each group, the LLM performs iterative multi-turn refinement using accumulated performance feedback, enabling related heuristics to be jointly analyzed and progressively improved across evolutionary iterations. Experiments on standard combinatorial optimization benchmarks show that ReVEL generally improves optimization performance over existing LLM-guided evolutionary baselines across multiple settings and LLM backbones. Additional analyses suggest that behavior-aware grouping contributes to more consistent refinement trajectories during iterative heuristic evolution. --- ### ReVEL: Multi-Turn Reflective LLM-Guided Heuristic Evolution via Structured Performance Feedback Authors: Cuong Van Duc, Minh Nguyen Dinh Tuan, Tam Vu Duc, Tung Vu Duy, Son Nguyen Van, Hanh Nguyen Thi, Binh Huynh Thi Thanh Published: 2026-03-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.04940v2 Designing effective heuristics for NP-hard combinatorial optimization problems remains challenging and often requires substantial domain expertise. Recent LLM-guided evolutionary methods have shown promise for automated heuristic generation, but most existing approaches refine heuristics independently or through limited pairwise feedback. We propose ReVEL: Multi-Turn Reflective LLM-Guided Heuristic Evolution via Structured Performance Feedback, a framework for group-wise multi-turn heuristic refinement. ReVEL organizes heuristics into behavior-aware reflective groups, including similarity-driven groups for localized refinement and diversity-driven groups for exploratory search. Within each group, the LLM performs iterative multi-turn refinement using accumulated performance feedback, enabling related heuristics to be jointly analyzed and progressively improved across evolutionary iterations. Experiments on standard combinatorial optimization benchmarks show that ReVEL generally improves optimization performance over existing LLM-guided evolutionary baselines across multiple settings and LLM backbones. Additional analyses suggest that behavior-aware grouping contributes to more consistent refinement trajectories during iterative heuristic evolution. --- ### DepthCharge: A Domain-Agnostic Framework for Measuring Depth-Dependent Knowledge in Large Language Models Authors: Alexander Sheppert Published: 2026-03-05 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.23514v1 Large Language Models appear competent when answering general questions but often fail when pushed into domain-specific details. No existing methodology provides an out-of-the-box solution for measuring how deeply LLMs can sustain accurate responses under adaptive follow-up questioning across arbitrary domains. We present DepthCharge, a domain-agnostic framework that measures knowledge depth through three innovations: adaptive probing that generates follow-up questions based on concepts the model actually mentions, on-demand fact verification from authoritative sources, and survival statistics with constant sample sizes at every depth level. The framework can be deployed on any knowledge domain with publicly verifiable facts, without requiring pre-constructed test sets or domain-specific expertise. DepthCharge results are relative to the evaluator model used for answer checking, making the framework a tool for comparative evaluation rather than absolute accuracy certification. Empirical validation across four diverse domains (Medicine, Constitutional Law, Ancient Rome, and Quantum Computing) with five frontier models demonstrates that DepthCharge reveals depth-dependent performance variation hidden by standard benchmarks. Expected Valid Depth (EVD) ranges from 3.45 to 7.55 across model-domain combinations, and model rankings vary substantially by domain, with no single model dominating all areas. Cost-performance analysis further reveals that expensive models do not always achieve deeper knowledge, suggesting that domain-specific evaluation is more informative than aggregate benchmarks for model selection in professional applications. --- ### Berta: an open-source, modular tool for AI-enabled clinical documentation Authors: Samridhi Vaid, Mike Weldon, Jesse Dunn, Sacha Davis, Kevin Lonergan, Henry Li, Jeffrey Franc, Mohamed Abdalla, Daniel C. Baumgart, Jake Hayward, J Ross Mitchell Published: 2026-03-05 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2603.23513v1 Commercial AI scribes cost \$99-600 per physician per month, operate as opaque systems, and do not return data to institutional infrastructure, limiting organizational control over data governance, quality improvement, and clinical workflows. We developed Berta, an open-source modular scribe platform for AI-enabled clinical documentation, and deployed a customized implementation within Alberta Health Services (AHS) integrated with their existing Snowflake AI Data Cloud infrastructure. The system combines automatic speech recognition with large language models while retaining all clinical data within the secure AHS environment. During eight months (November 2024 to July 2025), 198 emergency physicians used the system in 105 urban and rural facilities, generating 22148 clinical sessions and more than 2800 hours of audio. The use grew from 680 to 5530 monthly sessions. Operating costs averaged less than \$30 per physician per month, a 70-95% reduction compared to commercial alternatives. AHS has since approved expansion to 850 physicians. This is the first provincial-scale deployment of an AI scribe integrated with existing health system infrastructure. By releasing Berta as open source, we provide a reproducible, cost-effective alternative that health systems can adapt to their own secure environments, supporting data sovereignty and informed evaluation of AI documentation technology. --- ### S-Path-RAG: Semantic-Aware Shortest-Path Retrieval Augmented Generation for Multi-Hop Knowledge Graph Question Answering Authors: Rong Fu, Yemin Wang, Tianxiang Xu, Yongtai Liu, Weizhi Tang, Wangyu Wu, Xiaowen Ma, Simon Fong Published: 2026-03-05 Categories: cs.CL, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2603.23512v1 We present S-Path-RAG, a semantic-aware shortest-path Retrieval-Augmented Generation framework designed to improve multi-hop question answering over large knowledge graphs. S-Path-RAG departs from one-shot, text-heavy retrieval by enumerating bounded-length, semantically weighted candidate paths using a hybrid weighted $k$-shortest, beam, and constrained random-walk strategy, learning a differentiable path scorer together with a contrastive path encoder and lightweight verifier, and injecting a compact soft mixture of selected path latents into a language model via cross-attention. The system runs inside an iterative Neural-Socratic Graph Dialogue loop in which concise diagnostic messages produced by the language model are mapped to targeted graph edits or seed expansions, enabling adaptive retrieval when the model expresses uncertainty. This combination yields a retrieval mechanism that is both token-efficient and topology-aware while preserving interpretable path-level traces for diagnostics and intervention. We validate S-Path-RAG on standard multi-hop KGQA benchmarks and through ablations and diagnostic analyses. The results demonstrate consistent improvements in answer accuracy, evidence coverage, and end-to-end efficiency compared to strong graph- and LLM-based baselines. We further analyze trade-offs between semantic weighting, verifier filtering, and iterative updates, and report practical recommendations for deployment under constrained compute and token budgets. --- ### Compact Lifted Relaxations for Low-Rank Optimization Authors: Ryan Cory-Wright, Jean Pauphilet Published: 2026-03-05 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2603.20228v2 We develop tractable convex relaxations for rank-constrained quadratic optimization problems over $n \times m$ matrices, a setting for which tractable relaxations are typically only available when the objective or constraints admit spectral structure. We derive lifted semidefinite relaxations that do not require such spectral terms. Although a direct lifting introduces a large semidefinite constraint in dimension $n^2 + nm + 1$, we prove that many blocks of the moment matrix are redundant and derive an equivalent compact relaxation that only involves two semidefinite constraints of dimension $nm + 1$ and $n+m$, respectively. We also derive a new class of valid inequalities for low-rank problems, which we call projection cuts, that exploit the fact that rank constraints are inherited by linear images of a low-rank matrix, to strengthen our low-rank relaxations substantially. For matrix completion and reduced-rank regression problems, among others, we exploit additional structure to obtain even more compact formulations involving semidefinite matrices of dimension at most the sum of the two dimensions of the low-rank decision matrix (i.e., of size at most $n+m$). Overall, we obtain scalable semidefinite bounds for a broad class of low-rank quadratic problems. --- ### Physically Accurate Differentiable Inverse Rendering for Radio Frequency Digital Twin Authors: Xingyu Chen, Xinyu Zhang, Kai Zheng, Xinmin Fang, Tzu-Mao Li, Chris Xiaoxuan Lu, Zhengxiong Li Published: 2026-03-05 Categories: eess.SP, cs.GR, cs.LG arXiv: https://arxiv.org/abs/2603.18026v1 Digital twins, virtual simulated replicas of physical scenes, are transforming system design across industries. However, their potential in radio frequency (RF) systems has been limited by the non-differentiable nature of conventional RF simulators. The visibility of propagation paths causes severe discontinuities, and differentiable rendering techniques from computer graphics cannot easily transfer due to point-source antennas and dominant specular reflections. In this paper, we present RFDT, a physically based differentiable RF simulation framework that enables gradient-based interaction between virtual and physical worlds. RFDT resolves discontinuities with a physically grounded edge-diffraction transition function, and mitigates non-convexity from Fourier-domain processing through a signal domain transform surrogate. Our implementation demonstrates RFDT's ability to accurately reconstruct digital twins from real RF measurements. Moreover, RFDT can augment diverse downstream applications, such as test-time adaptation of machine learning-based RF sensing and physically constrained optimization of communication systems. --- ### Understanding the Relationship Between Firms' AI Technology Innovation and Consumer Complaints Authors: Yongchao Martin Ma, Zhongzhun Deng Published: 2026-03-05 Categories: cs.CY, cs.AI, stat.AP arXiv: https://arxiv.org/abs/2603.18025v1 In the artificial intelligence (AI) age, firms increasingly invest in AI technology innovation to secure competitive advantages. However, the relationship between firms' AI technology innovation and consumer complaints remains insufficiently explored. Drawing on Protection Motivation Theory (PMT), this paper investigates how firms' AI technology innovation influences consumer complaints. Employing a multimethod approach, Study 1 analyzes panel data from S&P 500 firms (N = 2,758 firm-year observations), Study 2 examines user-generated Reddit data (N = 2,033,814 submissions and comments), and Study 3 involves two controlled experiments (N = 410 and N = 500). The results reveal that firms' AI technology innovation significantly increases consumers' threat-related emotions, heightening their complaints. Furthermore, compared to AI process innovation, AI product innovation leads to higher consumer complaints. This paper advances the understanding of consumers' psychological responses to firms' AI innovation and provides practical implications for managing consumer complaints effectively. --- ### ProKWS: Personalized Keyword Spotting via Collaborative Learning of Phonemes and Prosody Authors: Jianan Pan, Yuanming Zhang, Kejie Huang Published: 2026-03-05 Categories: eess.AS, cs.AI, cs.CL, cs.SD arXiv: https://arxiv.org/abs/2603.18024v1 Current keyword spotting systems primarily use phoneme-level matching to distinguish confusable words but ignore user-specific pronunciation traits like prosody (intonation, stress, rhythm). This paper presents ProKWS, a novel framework integrating fine-grained phoneme learning with personalized prosody modeling. We design a dual-stream encoder where one stream derives robust phonemic representations through contrastive learning, while the other extracts speaker-specific prosodic patterns. A collaborative fusion module dynamically combines phonemic and prosodic information, enhancing adaptability across acoustic environments. Experiments show ProKWS delivers highly competitive performance, comparable to state-of-the-art models on standard benchmarks and demonstrates strong robustness for personalized keywords with tone and intent variations. --- ### PCOV-KWS: Multi-task Learning for Personalized Customizable Open Vocabulary Keyword Spotting Authors: Jianan Pan, Kejie Huang Published: 2026-03-05 Categories: eess.AS, cs.AI, cs.CL, cs.SD arXiv: https://arxiv.org/abs/2603.18023v1 As advancements in technologies like Internet of Things (IoT), Automatic Speech Recognition (ASR), Speaker Verification (SV), and Text-to-Speech (TTS) lead to increased usage of intelligent voice assistants, the demand for privacy and personalization has escalated. In this paper, we introduce a multi-task learning framework for personalized, customizable open-vocabulary Keyword Spotting (PCOV-KWS). This framework employs a lightweight network to simultaneously perform Keyword Spotting (KWS) and SV to address personalized KWS requirements. We have integrated a training criterion distinct from softmax-based loss, transforming multi-class classification into multiple binary classifications, which eliminates inter-category competition, while an optimization strategy for multi-task loss weighting is employed during training. We evaluated our PCOV-KWS system in multiple datasets, demonstrating that it outperforms the baselines in evaluation results, while also requiring fewer parameters and lower computational resources. --- ### Using Laplace Transform To Optimize the Hallucination of Generation Models Authors: Cheng Kang, Xinye Chen, Daniel Novak, Xujing Yao Published: 2026-03-05 Categories: math.OC, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2603.18022v1 To explore the feasibility of avoiding the confident error (or hallucination) of generation models (GMs), we formalise the system of GMs as a class of stochastic dynamical systems through the lens of control theory. Numerous factors can be attributed to the hallucination of the learning process of GMs, utilising knowledge of control theory allows us to analyse their system functions and system responses. Due to the high complexity of GMs when using various optimization methods, we cannot figure out their solution of Laplace transform, but from a macroscopic perspective, simulating the source response provides a virtual way to address the hallucination of GMs. We also find that the training progress is consistent with the corresponding system response, which offers us a useful way to develop a better optimization component. Finally, the hallucination problem of GMs is fundamentally optimized by using Laplace transform analysis. --- ### Quantum Key Distribution Secured Federated Learning for Channel Estimation and Radar Spectrum Sensing in 6G Networks Authors: Ferhat Ozgur Catak, Murat Kuzlu, Jungwon Seo, Umit Cali Published: 2026-03-05 Categories: cs.CR, cs.IT, cs.LG arXiv: https://arxiv.org/abs/2603.15649v1 This paper presents a federated learning framework secured by quantum key distribution (QKD) for wireless channel estimation and radar spectrum sensing in the next generation networks (NextG or Beyond 6G). A BB84-style protocol abstraction and pairwise additive masking are utilized to train clients' local models (CNN for channel estimation, U-Net for radar segmentation) and upload only masked model updates. The server aggregates without observing plain parameters; an eavesdropper without QKD keys cannot recover individual updates. Experiments show that secure FL achieves NMSE of 0.216 for channel estimation and 92.1\% accuracy with 0.72 mIoU for radar sensing. When an eavesdropper is present, QBER rises to $\sim$25\% and all rounds abort as intended; reconstruction error remains below $10^{-5}$, confirming correct aggregation. --- ### The Norm-Separation Delay Law of Grokking: A First-Principles Theory of Delayed Generalization Authors: Truong Xuan Khanh, Truong Quynh Hoa, Luu Duc Trung, Phan Thanh Duc Published: 2026-03-05 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.13331v2 Grokking -- the sudden generalisation that appears long after a model has perfectly memorised its training data -- has been widely observed but lacks a quantitative theory explaining the length of the delay. We show that grokking is a norm-driven representational phase transition in regularised training dynamics, and establish the Norm-Separation Delay Law: $T_{\mathrm{grok}} - T_{\mathrm{mem}} = Θ(γ_{\mathrm{eff}}^{-1} \log(\|θ_{\mathrm{mem}}\|^2 / \|θ_{\mathrm{post}}\|^2))$, where $γ_{\mathrm{eff}}$ is the optimiser's effective contraction rate ($γ_{\mathrm{eff}} = ηλ$ for SGD, $γ_{\mathrm{eff}} \ge ηλ$ for AdamW). The upper bound follows from a discrete Lyapunov contraction argument; the matching lower bound from dynamical constraints of regularised first-order optimisation. Across 293 training runs spanning modular addition, modular multiplication, and sparse parity, we confirm three falsifiable predictions: inverse scaling with weight decay ($R^2 = 0.97$), inverse scaling with learning rate ($R^2 = 0.92$), and logarithmic dependence on the norm ratio (Pearson $r = 0.91$). A fourth finding reveals that grokking requires an optimiser capable of decoupling memorisation from contraction: SGD fails entirely at the same hyperparameters where AdamW reliably groks. These results reframe grokking not as a mysterious optimisation artefact but as a predictable consequence of norm separation between competing interpolating representations. We further derive a practical three-input algorithm that predicts grokking delay at memorisation time with 34.6% mean absolute error (bootstrap 95% CI (30.0%, 39.4%), $N=60$ seeds), enabling principled early stopping. --- ### RBF-Solver: A Multistep Sampler for Diffusion Probabilistic Models via Radial Basis Functions Authors: Soochul Park, Yeon Ju Lee, SeongJin Yoon, Jiyub Shin, Juhee Lee, Seongwoon Jo Published: 2026-03-05 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2603.13330v1 Diffusion probabilistic models (DPMs) are widely adopted for their outstanding generative fidelity, yet their sampling is computationally demanding. Polynomial-based multistep samplers mitigate this cost by accelerating inference; however, despite their theoretical accuracy guarantees, they generate the sampling trajectory according to a predefined scheme, providing no flexibility for further optimization. To address this limitation, we propose RBF-Solver, a multistep diffusion sampler that interpolates model evaluations with Gaussian radial basis functions (RBFs). By leveraging learnable shape parameters in Gaussian RBFs, RBF-Solver explicitly follows optimal sampling trajectories. At first order, it reduces to the Euler method (DDIM). At second order or higher, as the shape parameters approach infinity, RBF-Solver converges to the Adams method, ensuring its compatibility with existing samplers. Owing to the locality of Gaussian RBFs, RBF-Solver maintains high image fidelity even at fourth order or higher, where previous samplers deteriorate. For unconditional generation, RBF-Solver consistently outperforms polynomial-based samplers in the high-NFE regime (NFE >= 15). On CIFAR-10 with the Score-SDE model, it achieves an FID of 2.87 with 15 function evaluations and further improves to 2.48 with 40 function evaluations. For conditional ImageNet 256 x 256 generation with the Guided Diffusion model at a guidance scale 8.0, substantial gains are achieved in the low-NFE range (5-10), yielding a 16.12-33.73% reduction in FID relative to polynomial-based samplers. --- ### LUMINA: Laplacian-Unifying Mechanism for Interpretable Neurodevelopmental Analysis via Quad-Stream GCN Authors: Minkyung Cha, Jooyoung Bae, Jaewon Jung, Ping Shu Ho, Ka Chun Cheung, Namjoon Kim Published: 2026-03-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.13329v1 Functional Magnetic Resonance Imaging(fMRI) has now become a classic way for measuring brain activity, and recent trend is shifting toward utilizing fMRI brain data for AI-driven diagnosis. Given that the brain functions as not a discrete but interconnected whole, Graph-based architectures represented by Graph Convolutional Network(GCN) has emerged as a dominant framework for such task, since they are capable of treating ROIs as dynamically interconnected nodes and extracting relational architecture between them. Ironically, however, it is the very nature of GCN's architecture that acts as an obstacle to its performance. The mathematical foundation of GCN, effective for capturing global regularities, acts as a tradeoff; by smoothing features across the connected nodes repeatedly, traditional GCN tend to blur out the contrastive dynamics that might be crucial in identifying certain neurological disorders. In order to break through this structural bottleneck, we propose LUMINA, a Laplacian-Unifying Mechanism for Interpretable Neurodevelopmental Analysis. Our model is a Quad-Stream GCN that employs a bipolar RELU activation and a dual-spectrum graph Laplacian filtering mechanism, thereby capturing heterogeneous dynamics that were often blurred out in conventional GCN. By doing so, we can preserve the diverse range and characteristics of neural connections in each fMRI data. Through 5-fold cross validation on the ADHD200(N=144) and ABIDE(N=579) dataset, LUMINA demonstrates stable diagnostic performance in two of the most critical neurodevelopmental disorder in childhood, ADHD and ASD, outperforming existing models with an accuracy of 84.66% and 88.41% each. --- ### Tureis: Transformer-based Unified Resilience for IoT Devices in Smart Homes Authors: Alireza Borhani, Vafa Andalibi, Bahar Asgari Published: 2026-03-05 Categories: cs.NI, cs.LG arXiv: https://arxiv.org/abs/2603.10038v1 Smart-home IoT systems rely on heterogeneous sensor networks whose correctness shapes application behavior and the physical environment. However, these low-cost, resource-constrained sensors are highly prone to failure under real-world stressors. Prior methods often assume single-failure, single-resident settings, offer only failure detection rather than sensor-level localization, cover limited fault types and sensor modalities, require labels and human intervention, or impose overheads hindering edge deployment. To overcome these limitations, we propose Tureis, a self-supervised, context-aware method for failure detection and faulty-sensor localization in smart homes, designed for multi-failure, multi-resident edge settings. Tureis encodes heterogeneous binary and numeric sensor streams into compact bit-level features. It then trains a lightweight BERT-style Transformer with sensor-wise masked reconstruction over short-horizon windows, capturing spatial and short-term temporal correlations without mixing unrelated events. This self-supervised objective removes the need for labels or curated semantics. Then, at run-time, Tureis converts reconstruction residuals into sensor-level failure evidence and uses an iterative isolate-and-continue loop that masks flagged sensors, allowing other failures to surface and enabling resilient, fine-grained localization. Across five datasets with up to nine residents, Tureis improves single-failure localization F1 by +7.6%, +21.0%, and +25.0% over three strong baselines. In multi-failure scenarios with up to five faulty sensors, it further boosts localization F1 by +17.6% and +35.4% over two baselines, while the third does not extend to this setting. These gains come with minute-scale localization and an edge-friendly footprint, as a sub-megabyte model that processes each minute of data in a few milliseconds with ~0.5 GB peak memory on a Raspberry Pi 5. --- ### Diagnosing FP4 inference: a layer-wise and block-wise sensitivity analysis of NVFP4 and MXFP4 Authors: Musa Cim, Burak Topcu, Mahmut Taylan Kandemir Published: 2026-03-05 Categories: cs.AR, cs.AI arXiv: https://arxiv.org/abs/2603.08747v1 Quantization addresses the high resource demand for large language models (LLMs) by alleviating memory pressure and bandwidth congestion and providing significantly scaled compute power with a tolerable impact on accuracy. Four-bit floating point (FP4), the lowest-precision format that preserves essential numerical properties such as exponent and sign, has begun to be adopted in cutting-edge architectures, including Blackwell and AMD CDNA, to support LLM quantization and reduce deployment costs. Although aggressive quantization can yield efficiency gains, the quantization sensitivity of within-transformer layers and whether these sensitivities generalize across existing FP4 formats and model scales remain underexplored. To elucidate quantization sensitivity, this study conducts a systematic analysis of two FP4 formats, MXFP4 and NVFP4, across three Qwen2.5 model scales (0.5B, 7B, and 14B), using controlled component-wise and block-wise isolation methodologies. We observe that MLP up- and down-projection layers consistently dominate in terms of sensitivity, while gate and attention projections are moderately and substantially less sensitive to FP4 quantization, respectively. We further find that sensitivity does not universally localize to the final blocks, but early blocks can be highly sensitive, particularly under MXFP4. Our results provide a diagnostic characterization of the inference behavior of FP4 across components, depths, and FP4 formats. --- ### Dynamic Targeting of Satellite Observations Using Supplemental Geostationary Satellite Data and Hierarchical Planning Authors: Akseli Kangaslahti, Itai Zilberstein, Alberto Candela, Steve Chien Published: 2026-03-05 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2603.06719v1 The Dynamic Targeting (DT) mission concept is an approach to satellite observation in which a lookahead sensor gathers information about the upcoming environment and uses this information to intelligently plan observations. Previous work has shown that DT has the potential to increase the science return across applications. However, DT mission concepts must address challenges, such as the limited spatial extent of onboard lookahead data and instrument mobility, data throughput, and onboard computation constraints. In this work, we show how the performance of DT systems can be improved by using supplementary data streamed from geostationary satellites that provide lookahead information up to 35 minutes ahead of time rather than the 1 minute latency from an onboard lookahead sensor. While there is a greater volume of geostationary data, the search space for observation planning explodes exponentially with the size of the horizon. To address this, we introduce a hierarchical planning approach in which the geostationary data is used to plan a long-term observation blueprint in polynomial time, then the onboard lookahead data is leveraged to refine that plan over short-term horizons. We compare the performance of our approach to that of traditional DT planners relying on onboard lookahead data across four different problem instances: three cloud avoidance variations and a storm hunting scenario. We show that our hierarchical planner outperforms the traditional DT planners by up to 41% and examine the features of the scenarios that affect the performance of our approach. We demonstrate that incorporating geostationary satellite data is most effective for dynamic problem instances in which the targets of interest are sparsely distributed throughout the overflight. --- ### Scaling Agentic Capabilities, Not Context: Efficient Reinforcement Finetuning for Large Toolspaces Authors: Karan Gupta, Pranav Vajreshwari, Yash Pandya, Raghav Magazine, Akshay Nambi, Ahmed Awadallah Published: 2026-03-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.06713v1 Agentic systems operating over large tool ecosystems must plan and execute long-horizon workflows under weak or non-verifiable supervision. While frontier models mitigate these challenges through scale and large context budgets, small language models (SLMs) remain brittle: eager tool loading saturates context, execution errors compound over time, and sparse rewards limit learning. We introduce ATLAS, a reinforcement finetuning framework that enables SLMs to operate effectively in large-scale toolspace environments by learning how to acquire context and how to execute actions. Our approach makes two key contributions. First, we treat context control and execution structure as learnable decisions, combining iterative tool loading with programmatic tool orchestration to bound context growth and stabilize long-horizon trajectories. Second, we propose rubric-based reinforcement finetuning, which decomposes task success into structured, task-aligned criteria and enables scalable training using small judge models. Across MCP benchmarks, these design choices yield large and consistent gains over generic RL baselines, allowing a 4B SLM to approach frontier-agent performance under far tighter parameter and context budgets. --- ### Uncertainty-Aware Solar Flare Regression Authors: Jinsu Hong, Chetraj Pandey, Berkay Aydin Published: 2026-03-05 Categories: astro-ph.SR, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2603.06712v1 Current solar flare predictions often lack precise quantification of their reliability, resulting in frequent false alarms, particularly when dealing with datasets skewed towards extreme events. To improve the trustworthiness of space weather forecasting, it is crucial to establish confidence intervals for model predictions. Conformal prediction, a machine learning framework, presents a promising avenue for this purpose by constructing prediction intervals that ensure valid coverage in finite samples without making assumptions about the underlying data distribution. In this study, we explore the application of conformal prediction to regression tasks in space weather forecasting. Specifically, we implement full-disk solar flare prediction using images created from magnetic field maps and adapt four pre-trained deep learning models to incorporate three distinct methods for constructing confidence intervals: conformal prediction, quantile regression, and conformalized quantile regression. Our experiments demonstrate that conformalized quantile regression achieves higher coverage rates and more favorable average interval lengths compared to alternative methods, underscoring its effectiveness in enhancing the reliability of solar weather forecasting models. --- ### Mining Beyond the Bools: Learning Data Transformations and Temporal Specifications Authors: Sam Nicholas Kouteili, William Fishell, Christian Scaff, Mark Santolucito, Ruzica Piskac Published: 2026-03-05 Categories: cs.LO, cs.AI, cs.FL, cs.PL arXiv: https://arxiv.org/abs/2603.06710v1 Mining specifications from execution traces presents an automated way of capturing characteristic system behaviors. However, existing approaches are largely restricted to Boolean abstractions of events, limiting their ability to express data-aware properties. In this paper, we extend mining procedures to operate over richer datatypes. We first establish candidate functions in our domain that cover the set of traces by leveraging Syntax Guided Synthesis (SyGuS) techniques. To capture these function applications temporally, we formalize the semantics of TSL$_f$, a finite-prefix interpretation of Temporal Stream Logic (TSL) that extends LTL$_f$ with support for first-order predicates and functional updates. This allows us to unify a corresponding procedure for learning the data transformations and temporal specifications of a system. We demonstrate our approach synthesizing reactive programs from mined specifications on the OpenAI-Gymnasium ToyText environments, finding that our method is more robust and orders of magnitude more sample-efficient than passive learning baselines on generalized problem instances. --- ### On the Generalization Capacities of MLLMs for Spatial Intelligence Authors: Gongjie Zhang, Wenhao Li, Quanhao Qian, Jiuniu Wang, Deli Zhao, Shijian Lu, Ran Xu Published: 2026-03-05 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2603.06704v1 Multimodal Large Language Models (MLLMs) that directly process RGB inputs for tasks like 3D localization and navigation have shown remarkable potential. However, we argue that these RGB-only approaches are fundamentally flawed in their ability to generalize across cameras. By ignoring camera parameters, they entangle an object's physical properties with the camera's perspective, creating an irresolvable ambiguity. We show this leads MLLMs to overfit to the training camera distribution, rather than learning true and generalizable 3D geometric principles. To address this, we propose Camera-Aware MLLM framework for spatial MLLMs. It learns generalizable spatial reasoning by: (i) injecting camera intrinsics via a dense embedding that conditions each visual token; (ii) introducing a camera-aware data augmentation strategy that synthetically varies camera parameters, forcing the model to disentangle camera properties from scene content; and (iii) distilling geometric priors from a 3D vision foundation model. Extensive experiments demonstrate that camera-aware MLLMs substantially outperform their naive counterparts, particularly in cross-camera generalization tests on spatially-grounded tasks, indicating that camera-awareness is not only beneficial but also a prerequisite for robust and generalizable spatial intelligence in MLLMs. --- ### SIQA: Toward Reliable Scientific Image Quality Assessment Authors: Wenzhe Li, Liang Chen, Junying Wang, Yijing Guo, Ye Shen, Farong Wen, Chunyi Li, Zicheng Zhang, Guangtao Zhai Published: 2026-03-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06700v1 Scientific images fundamentally differ from natural and AI-generated images in that they encode structured domain knowledge rather than merely depict visual scenes. Assessing their quality therefore requires evaluating not only perceptual fidelity but also scientific correctness and logical completeness. However, existing image quality assessment (IQA) paradigms primarily focus on perceptual distortions or image-text alignment, implicitly assuming that depicted content is factually valid. This assumption breaks down in scientific contexts, where visually plausible figures may still contain conceptual errors or incomplete reasoning. To address this gap, we introduce Scientific Image Quality Assessment (SIQA), a framework that models scientific image quality along two complementary dimensions: Knowledge (Scientific Validity and Scientific Completeness) and Perception (Cognitive Clarity and Disciplinary Conformity). To operationalize this formulation, we design two evaluation protocols: SIQA-U (Understanding), which measures semantic comprehension of scientific content through multiple-choice tasks, and SIQA-S (Scoring), which evaluates alignment with expert quality judgments. We further construct the SIQA Challenge, consisting of an expert-annotated benchmark and a large-scale training set. Experiments across representative multimodal large language models (MLLMs) reveal a consistent discrepancy between scoring alignment and scientific understanding. While models can achieve strong agreement with expert ratings under SIQA-S, their performance on SIQA-U remains substantially lower. Fine-tuning improves both metrics, yet gains in scoring consistently outpace improvements in understanding. These results suggest that rating consistency alone may not reliably reflect scientific comprehension, underscoring the necessity of multidimensional evaluation for scientific image quality assessment. --- ### Multi-label Instance-level Generalised Visual Grounding in Agriculture Authors: Mohammadreza Haghighat, Alzayat Saleh, Mostafa Rahimi Azghadi Published: 2026-03-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06699v1 Understanding field imagery such as detecting plants and distinguishing individual crop and weed instances is a central challenge in precision agriculture. Despite progress in vision-language tasks like captioning and visual question answering, Visual Grounding (VG), localising language-referred objects, remains unexplored in agriculture. A key reason is the lack of suitable benchmark datasets for evaluating grounding models in field conditions, where many plants look highly similar, appear at multiple scales, and the referred target may be absent from the image. To address these limitations, we introduce gRef-CW, the first dataset designed for generalised visual grounding in agriculture, including negative expressions. Benchmarking current state-of-the-art grounding models on gRef-CW reveals a substantial domain gap, highlighting their inability to ground instances of crops and weeds. Motivated by these findings, we introduce Weed-VG, a modular framework that incorporates multi-label hierarchical relevance scoring and interpolation-driven regression. Weed-VG advances instance-level visual grounding and provides a clear baseline for developing VG methods in precision agriculture. Code will be released upon acceptance. --- ### Breaking the Geometric Bottleneck: Contrastive Expansion in Asymmetric Cross-Modal Distillation Authors: Kabir Thayani Published: 2026-03-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06698v3 Knowledge distillation between asymmetric architectures often induces severe geometric constraints on the learned representation space. We investigate dimensional collapse when distilling global Vision Transformers into capacity-constrained, local-receptive-field CNNs (0.5M-8.0M parameters). Using strictly centered SVD and Shannon Entropy Effective Rank, we confirm capacity-agnostic collapse under cosine distillation: a CLIP ViT-B/32 Teacher exhibits Effective Rank 88.68 on CIFAR-10, while all cosine-distilled students collapse to ~17 regardless of parameter count. An auxiliary InfoNCE objective expands this to ~41 dimensions. Critically, we ask whether this expansion is functionally useful. Multi-seed linear-probe evaluation shows InfoNCE expansion degrades downstream accuracy by 15-18 points relative to the collapsed baseline, despite more than doubling Effective Rank. A class-structure decomposition traces this to signal dilution: InfoNCE's class-blind uniformity pressure weakens class-discriminative structure in the original dimensions while adding only weakly relevant structure elsewhere. We then test a label-aware alternative, Supervised Contrastive distillation. On CIFAR-100, where we swept student capacity directly, its Effective Rank is invariant to capacity; on CIFAR-10, at a single tested width, it settles to a lower rank while matching baseline accuracy. Sweeping temperature instead of capacity, rank and downstream accuracy increase together monotonically. These results show Effective Rank alone is not a reliable proxy for representation quality: whether expansion helps or harms downstream performance depends on whether the driving objective is label-aware, not the magnitude of expansion itself. --- ### Breaking the Geometric Bottleneck: Contrastive Expansion in Asymmetric Cross-Modal Distillation Authors: Kabir Thayani Published: 2026-03-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06698v2 Knowledge distillation between asymmetric architectures often induces severe geometric constraints on the learned representation space. In this work, we investigate the Dimensional Collapse phenomenon when distilling global Vision Transformers (CLIP and DINOv2) into capacity-constrained CNNs. By employing strictly centered SVD and Effective Rank, we first demonstrate a capacity-agnostic phase transition on CIFAR-10 where standard cosine distillation collapses representations to an intrinsic Effective Rank of ~16. To reverse this, we integrate an auxiliary contrastive objective (InfoNCE), expanding the student's manifold by 2.4x (to ~38 effective dimensions). We further demonstrate that while DINOv2's uniform geometry partially prevents collapse, contrastive expansion remains a universal requirement to reach the CNN's topological capacity limit (~82 dimensions). Finally, we reveal a critical capacity-density trade-off: overparameterization within fixed manifolds induces brittleness, while capacity-constrained models act as optimal low-pass semantic filters, successfully recovering inherent noise immunity. --- ### Thinking with Gaze: Sequential Eye-Tracking as Visual Reasoning Supervision for Medical VLMs Authors: Yiwei Li, Yifan Zhou, Huaqin Zhao, Zihao Wu, Zhengliang Liu, Xiang Li, Quanzheng Li, Tianming Liu, Lin Zhao Published: 2026-03-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.06697v2 Vision--language models (VLMs) process images as visual tokens, yet their intermediate reasoning is often carried out in text, which can be suboptimal for visually grounded radiology tasks. Radiologists instead diagnose via sequential visual search; eye-tracking captures this process as time-ordered gaze trajectories that reveal how evidence is acquired over time. We use eye-gaze as supervision to guide VLM reasoning by introducing a small set of dedicated gaze tokens. These tokens are trained to predict gaze-selected image patch indices in temporal order, encouraging the model to follow human-like evidence acquisition and integration. Experiments on MIMIC-EYE and multiple external zero-shot benchmarks show consistent gains over baselines, achieving state-of-the-art in-domain performance and improved out-of-domain robustness. These results highlight temporally ordered gaze as an effective supervision signal for learning visually grounded medical reasoning. --- ### Thinking with Gaze: Sequential Eye-Tracking as Visual Reasoning Supervision for Medical VLMs Authors: Yiwei Li, Zihao Wu, Yanjun Lv, Hanqi Jiang, Weihang You, Zhengliang Liu, Dajiang Zhu, Xiang Li, Quanzheng Li, Tianming Liu, Lin Zhao Published: 2026-03-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.06697v1 Vision--language models (VLMs) process images as visual tokens, yet their intermediate reasoning is often carried out in text, which can be suboptimal for visually grounded radiology tasks. Radiologists instead diagnose via sequential visual search; eye-tracking captures this process as time-ordered gaze trajectories that reveal how evidence is acquired over time. We use eye-gaze as supervision to guide VLM reasoning by introducing a small set of dedicated gaze tokens. These tokens are trained to predict gaze-selected image patch indices in temporal order, encouraging the model to follow human-like evidence acquisition and integration. Experiments on MIMIC-EYE and multiple external zero-shot benchmarks show consistent gains over baselines, achieving state-of-the-art in-domain performance and improved out-of-domain robustness. These results highlight temporally ordered gaze as an effective supervision signal for learning visually grounded medical reasoning. --- ### HARP: HARmonizing in-vivo diffusion MRI using Phantom-only training Authors: Hwihun Jeong, Qiang Liu, Kathryn E. Keenan, Elisabeth A. Wilde, Walter Schneider, Sudhir Pathak, Anthony Zuccolotto, Lauren J. O'Donnell, Lipeng Ning, Yogesh Rathi Published: 2026-03-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.06696v1 Purpose: Combining multi-site diffusion MRI (dMRI) data is hindered by inter-scanner variability, which confounds subsequent analysis. Previous harmonization methods require large, matched or traveling human subjects from multiple sites, which are impractical to acquire in many situations. This study aims to develop a deep learning-based dMRI harmonization framework that eliminates the reliance on multi-site in-vivo traveling human data for training. Methods: HARP employs a voxel-wise 1D neural network trained on an easily transportable diffusion phantom. The model learns relationships between spherical harmonics coefficients of different sites without memorizing spatial structures. Results: HARP reduced inter-scanner variability levels significantly in various measures. Quantitatively, it decreased inter-scanner variability as measured by standard error in FA (12%), MD (10%), and GFA (30%) with scan-rescan standard error as the baseline, while preserving fiber orientations and tractography after harmonization. Conclusion: We believe that HARP represents an important first step toward dMRI harmonization using only phantom data, thereby obviating the need for complex, matched in vivo multi-site cohorts. This phantom-only strategy substantially enhances the feasibility and scalability of quantitative dMRI for large-scale clinical studies. --- ### Predicting Atomistic Transitions with Transformers Authors: Henry Tischler, Wenting Li, Qi Tang, Danny Perez, Thomas Vogel Published: 2026-03-05 Categories: cond-mat.mtrl-sci, cs.LG arXiv: https://arxiv.org/abs/2603.06526v2 Accurate knowledge of the atomistic transition pathways in materials and material surfaces is crucial for many material science problems. However, conventional simulation techniques used to find these transitions are extremely computationally intensive. Even with large-scale, accelerated material simulations, the computational cost constrains the applicable domain in practice. Machine learning models, with the potential to learn the complex emergent behaviors governing atomistic transitions as a fast surrogate model, have great promise to predict transitions with a vastly reduced computational cost. Here, we demonstrate how transformers can be trained to predict atomistic transitions in nano-clusters. We show how we evaluate physical validity of the predictions and how a multitude of additional, different microstates can be generated by slightly varying the data provided to the model. --- ### TML-Bench: Benchmark for Data Science Agents on Tabular ML Tasks Authors: Mykola Pinchuk Published: 2026-03-05 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.05764v1 Autonomous coding agents can produce strong tabular baselines quickly on Kaggle-style tasks. Practical value depends on end-to-end correctness and reliability under time limits. This paper introduces TML-Bench, a tabular benchmark for data science agents on Kaggle-style tasks. This paper evaluates 10 OSS LLMs on four Kaggle competitions and three time budgets (240s, 600s, and 1200s). Each model is run five times per task and budget. A run is successful if it produces a valid submission and a private-holdout score on hidden labels that are not accessible to the agent. This paper reports median performance, success rates, and run-to-run variability. MiniMax-M2.1 model achieves the best aggregate performance score on all four competitions under the paper's primary aggregation. Average performance improves with larger time budgets. Scaling is noisy for some individual models at the current run count. Code and materials are available at https://github.com/MykolaPinchuk/TML-bench/tree/master. --- ### Score-Guided Proximal Projection: A Unified Geometric Framework for Rectified Flow Editing Authors: Vansh Bansal, James G Scott Published: 2026-03-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.05761v1 Rectified Flow (RF) models achieve state-of-the-art generation quality, yet controlling them for precise tasks -- such as semantic editing or blind image recovery -- remains a challenge. Current approaches bifurcate into inversion-based guidance, which suffers from "geometric locking" by rigidly adhering to the source trajectory, and posterior sampling approximations (e.g., DPS), which are computationally expensive and unstable. In this work, we propose Score-Guided Proximal Projection (SGPP), a unified framework that bridges the gap between deterministic optimization and stochastic sampling. We reformulate the recovery task as a proximal optimization problem, defining an energy landscape that balances fidelity to the input with realism from the pre-trained score field. We theoretically prove that this objective induces a normal contraction property, geometrically guaranteeing that out-of-distribution inputs are snapped onto the data manifold, and it effectively reaches the posterior mode constrained to the manifold. Crucially, we demonstrate that SGPP generalizes state-of-the-art editing methods: RF-inversion is effectively a limiting case of our framework. By relaxing the proximal variance, SGPP enables "soft guidance," offering a continuous, training-free trade-off between strict identity preservation and generative freedom. --- ### MIRACL: A Diverse Meta-Reinforcement Learning for Multi-Objective Multi-Echelon Combinatorial Supply Chain Optimisation Authors: Rifny Rachman, Josh Tingey, Richard Allmendinger, Wei Pan, Pradyumn Shukla, Bahrul Ilmi Nasution Published: 2026-03-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.05760v1 Multi-objective reinforcement learning (MORL) is effective for multi-echelon combinatorial supply chain optimisation, where tasks involve high dimensionality, uncertainty, and competing objectives. However, its deployment in dynamic environments is hindered by the need for task-specific retraining and substantial computational cost. We introduce MIRACL (Meta multI-objective Reinforcement leArning with Composite Learning), a hierarchical Meta-MORL framework that allows for a few-shot generalisation across diverse tasks. MIRACL decomposes each task into structured subproblems for efficient policy adaptation and meta-learns a global policy across tasks using a Pareto-based adaptation strategy to encourage diversity in meta-training and fine-tuning. To our knowledge, this is the first integration of Meta-MORL with such mechanisms in combinatorial optimisation. Although validated in the supply chain domain, MIRACL is theoretically domain-agnostic and applicable to broader dynamic multi-objective decision-making problems. Empirical evaluations show that MIRACL outperforms conventional MORL baselines in simple to moderate tasks, achieving up to 10% higher hypervolume and 5% better expected utility. These results underscore the potential of MIRACL for robust, efficient adaptation in multi-objective problems. --- ### Full Dynamic Range Sky-Modelling For Image Based Lighting Authors: Ian J. Maquignaz Published: 2026-03-05 Categories: cs.CV, cs.GR, cs.LG arXiv: https://arxiv.org/abs/2603.05758v1 Accurate environment maps are a key component to modelling real-world outdoor scenes. They enable captivating visual arts, immersive virtual reality and a wide range of scientific and engineering applications. To alleviate the burden of physical-capture, physically-simulation and volumetric rendering, sky-models have been proposed as fast, flexible, and cost-saving alternatives. In recent years, sky-models have been extended through deep learning to be more comprehensive and inclusive of cloud formations, but recent work has demonstrated these models fall short in faithfully recreating accurate and photorealistic natural skies. Particularly at higher resolutions, DNN sky-models struggle to accurately model the 14EV+ class-imbalanced solar region, resulting in poor visual quality and scenes illuminated with skewed light transmission, shadows and tones. In this work, we propose Icarus, an all-weather sky-model capable of learning the exposure range of Full Dynamic Range (FDR) physically captured outdoor imagery. Our model allows conditional generation of environment maps with intuitive user-positioning of solar and cloud formations, and extends on current state-of-the-art to enable user-controlled texturing of atmospheric formations. Through our evaluation, we demonstrate Icarus is interchangeable with FDR physically captured outdoor imagery or parametric sky-models, and illuminates scenes with unprecedented accuracy, photorealism, lighting directionality (shadows), and tones in Image Based Lightning (IBL). --- ### Uni-LVC: A Unified Method for Intra- and Inter-Mode Learned Video Compression Authors: Yichi Zhang, Ruoyu Yang, Fengqing Zhu Published: 2026-03-05 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2603.05756v1 Recent advances in learned video compression (LVC) have led to significant performance gains, with codecs such as DCVC-RT surpassing the H.266/VVC low-delay mode in compression efficiency. However, existing LVCs still exhibit key limitations: they often require separate models for intra and inter coding modes, and their performance degrades when temporal references are unreliable. To address this, we introduce Uni-LVC, a unified LVC method that supports both intra and inter coding with low-delay and random-access in a single model. Building on a strong intra-codec, Uni-LVC formulates inter-coding as intra-coding conditioned on temporal information extracted from reference frames. We design an efficient cross-attention adaptation module that integrates temporal cues, enabling seamless support for both unidirectional (low-delay) and bidirectional (random-access) prediction modes. A reliability-aware classifier is proposed to selectively scale the temporal cues, making Uni-LVC behave closer to intra coding when references are unreliable. We further propose a multistage training strategy to facilitate adaptive learning across various coding modes. Extensive experiments demonstrate that Uni-LVC achieves superior rate-distortion performance in intra and inter configurations while maintaining comparable computational efficiency. --- ### NERdME: a Named Entity Recognition Dataset for Indexing Research Artifacts in Code Repositories Authors: Genet Asefa Gesese, Zongxiong Chen, Shufan Jiang, Mary Ann Tan, Zhaotai Liu, Sonja Schimmler, Harald Sack Published: 2026-03-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.05750v1 Existing scholarly information extraction (SIE) datasets focus on scientific papers and overlook implementation-level details in code repositories. README files describe datasets, source code, and other implementation-level artifacts, however, their free-form Markdown offers little semantic structure, making automatic information extraction difficult. To address this gap, NERdME is introduced: 200 manually annotated README files with over 10,000 labeled spans and 10 entity types. Baseline results using large language models and fine-tuned transformers show clear differences between paperlevel and implementation-level entities, indicating the value of extending SIE benchmarks with entity types available in README files. A downstream entity-linking experiment was conducted to demonstrate that entities derived from READMEs can support artifact discovery and metadata integration. ---