# Research Papers — 2026-02 · bundle page 31 of 210 > Full-text markdown of papers #1501–1550 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/31 [← Previous bundle](https://aipod.dev/papers/archive/2026-02/page/30.md) [Next bundle →](https://aipod.dev/papers/archive/2026-02/page/32.md) ### Evaluating the Usage of African-American Vernacular English in Large Language Models Authors: Deja Dunlap, R. Thomas McCoy Published: 2026-02-25 Categories: cs.CL, cs.HC arXiv: https://arxiv.org/abs/2602.21485v1 In AI, most evaluations of natural language understanding tasks are conducted in standardized dialects such as Standard American English (SAE). In this work, we investigate how accurately large language models (LLMs) represent African American Vernacular English (AAVE). We analyze three LLMs to compare their usage of AAVE to the usage of humans who natively speak AAVE. We first analyzed interviews from the Corpus of Regional African American Language and TwitterAAE to identify the typical contexts where people use AAVE grammatical features such as ain't. We then prompted the LLMs to produce text in AAVE and compared the model-generated text to human usage patterns. We find that, in many cases, there are substantial differences between AAVE usage in LLMs and humans: LLMs usually underuse and misuse grammatical features characteristic of AAVE. Furthermore, through sentiment analysis and manual inspection, we found that the models replicated stereotypes about African Americans. These results highlight the need for more diversity in training data and the incorporation of fairness methods to mitigate the perpetuation of stereotypes. --- ### Unified Unsupervised and Sparsely-Supervised 3D Object Detection by Semantic Pseudo-Labeling and Prototype Learning Authors: Yushen He, Lei Zhao, Weidong Chen Published: 2026-02-25 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.21484v2 3D object detection is essential for autonomous driving and robotic perception, yet its reliance on large-scale manually annotated data limits scalability and adaptability. To reduce annotation dependency, unsupervised and sparsely-supervised paradigms have emerged. However, they face intertwined challenges: low-quality pseudo-labels, unstable feature mining, and a lack of a unified training framework. This paper proposes SPL, a unified training framework for both unsupervised and sparsely-supervised 3D object detection via \underline{S}emantic \underline{P}seudo-labeling and prototype \underline{L}earning. SPL first generates high-quality pseudo-labels by integrating image semantics, point cloud geometry, and temporal cues, producing both 3D bounding boxes for dense objects and 3D point labels for sparse ones. These pseudo-labels are not used directly but as probabilistic priors within a novel, multi-stage prototype learning strategy. This strategy stabilizes feature representation learning through memory-based initialization and momentum-based prototype updating, effectively mining features from both labeled and unlabeled data. Extensive experiments on KITTI and nuScenes datasets demonstrate that SPL significantly outperforms state-of-the-art methods in both settings. Our work provides a robust and generalizable solution for learning 3D object detectors with minimal or no manual annotations. Our code is available at https://github.com/TossherO/SPL. --- ### Perceptual Quality Optimization of Image Super-Resolution Authors: Wei Zhou, Yixiao Li, Hadi Amirpour, Xiaoshuai Hao, Jiang Liu, Peng Wang, Hantao Liu Published: 2026-02-25 Categories: eess.IV, cs.CV, cs.MM arXiv: https://arxiv.org/abs/2602.21482v1 Single-image super-resolution (SR) has achieved remarkable progress with deep learning, yet most approaches rely on distortion-oriented losses or heuristic perceptual priors, which often lead to a trade-off between fidelity and visual quality. To address this issue, we propose an \textit{Efficient Perceptual Bi-directional Attention Network (Efficient-PBAN)} that explicitly optimizes SR towards human-preferred quality. Unlike patch-based quality models, Efficient-PBAN avoids extensive patch sampling and enables efficient image-level perception. The proposed framework is trained on our self-constructed SR quality dataset that covers a wide range of state-of-the-art SR methods with corresponding human opinion scores. Using this dataset, Efficient-PBAN learns to predict perceptual quality in a way that correlates strongly with subjective judgments. The learned metric is further integrated into SR training as a differentiable perceptual loss, enabling closed-loop alignment between reconstruction and perceptual assessment. Extensive experiments demonstrate that our approach delivers superior perceptual quality. Code is publicly available at https://github.com/Lighting-YXLI/Efficient-PBAN. --- ### Both Ends Count! Just How Good are LLM Agents at "Text-to-Big SQL"? Authors: Germán T. Eizaguirre, Lars Tissen, Marc Sánchez-Artigas Published: 2026-02-25 Categories: cs.DB, cs.CL, cs.IR arXiv: https://arxiv.org/abs/2602.21480v4 Text-to-SQL and Big Data are both extensively benchmarked fields, yet there is limited research that evaluates them jointly. In the real world, Text-to-SQL systems are often embedded with Big Data workflows, such as large-scale data processing or interactive data analytics. We refer to this as ``Text-to-Big SQL''. However, existing text-to-SQL benchmarks remain narrowly scoped and overlook the cost and performance implications that arise at scale. For instance, translation errors that are minor on small datasets lead to substantial cost and latency overheads as data scales, a relevant issue completely ignored by text-to-SQL metrics. In this paper, we overcome this overlooked challenge by introducing novel and representative metrics for evaluating Text-to-Big SQL. Our study focuses on production-level LLM agents, a database-agnostic system adaptable to diverse user needs. Via an extensive evaluation of frontier models, we show that text-to-SQL metrics are insufficient for Big Data. In contrast, our proposed text-to-Big SQL metrics accurately reflect execution efficiency, cost, and the impact of data scale. For example, GPT-4o compensates for roughly 7% lower accuracy than the top-performing later-generation models with up to a 12.16x speedup, while GPT-5.2 is more than twice as cost-effective as Gemini 3 Pro at large input scales. --- ### Global Sequential Testing for Multi-Stream Auditing Authors: Beepul Bharti, Ambar Pal, Jeremias Sulam Published: 2026-02-25 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2602.21479v3 Across many risk-sensitive areas, it is critical to continuously audit machine learning systems as we receive more data to quickly determine if they are performing as designed. This auditing task can be modeled as a sequential hypothesis testing problem with $k$ data streams and a global null hypothesis that asserts the system operates as intended across all $k$ streams. Under the alternative, the standard global sequential test, which uses a Bonferroni correction, has an expected stopping time of $O\left(\ln \frac{k}α\right)$ for large $k$ and significance level $α$. In this work, we demonstrate that efficient sequential tests, relying on merging martingales via averaging and products rules, provide improved stopping times, and thus more powerful tests against the null. Using these results, we show that a balanced test can match the Bonferroni rate of $O\left(\ln \frac{k}α\right)$ in the sparse regime (just a few non-null streams) while achieving $O\left(\frac{1}{k}\ln \frac{1}α\right)$ under dense alternatives (many non-null steams). We validate our theory through experiments on both synthetic and real-world data. --- ### Efficient Inference after Directionally Stable Adaptive Experiments Authors: Zikai Shen, Houssam Zenati, Nathan Kallus, Arthur Gretton, Koulik Khamaru, Aurélien Bibaut Published: 2026-02-25 Categories: stat.ML, cs.LG, math.ST, stat.ME arXiv: https://arxiv.org/abs/2602.21478v1 We study inference on scalar-valued pathwise differentiable targets after adaptive data collection, such as a bandit algorithm. We introduce a novel target-specific condition, directional stability, which is strictly weaker than previously imposed target-agnostic stability conditions. Under directional stability, we show that estimators that would have been efficient under i.i.d. data remain asymptotically normal and semiparametrically efficient when computed from adaptively collected trajectories. The canonical gradient has a martingale form, and directional stability guarantees stabilization of its predictable quadratic variation, enabling high-dimensional asymptotic normality. We characterize efficiency using a convolution theorem for the adaptive-data setting, and give a condition under which the one-step estimator attains the efficiency bound. We verify directional stability for LinUCB, yielding the first semiparametric efficiency guarantee for a regular scalar target under LinUCB sampling. --- ### A Knowledge-Driven Approach to Music Segmentation, Music Source Separation and Cinematic Audio Source Separation Authors: Chun-wei Ho, Sabato Marco Siniscalchi, Kai Li, Chin-Hui Lee Published: 2026-02-25 Categories: eess.AS, cs.AI, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2602.21476v1 We propose a knowledge-driven, model-based approach to segmenting audio into single-category and mixed-category chunks with applications to source separation. "Knowledge" here denotes information associated with the data, such as music scores. "Model" here refers to tool that can be used for audio segmentation and recognition, such as hidden Markov models. In contrast to conventional learning that often relies on annotated data with given segment categories and their corresponding boundaries to guide the learning process, the proposed framework does not depend on any pre-segmented training data and learns directly from the input audio and its related knowledge sources to build all necessary models autonomously. Evaluation on simulation data shows that score-guided learning achieves very good music segmentation and separation results. Tested on movie track data for cinematic audio source separation also shows that utilizing sound category knowledge achieves better separation results than those obtained with data-driven techniques without using such information. --- ### Automatic Map Density Selection for Locally-Performant Visual Place Recognition Authors: Somayeh Hussaini, Tobias Fischer, Michael Milford Published: 2026-02-25 Categories: cs.CV arXiv: https://arxiv.org/abs/2602.21473v3 A key challenge in translating Visual Place Recognition (VPR) from the lab to long-term deployment is ensuring a priori that a system can meet user-specified performance requirements across different parts of an environment, rather than just on average globally. One critical mechanism for controlling this local performance is the density of the reference mapping database, yet this factor is largely neglected in existing work, where fixed, engineering-driven sampling densities based on sensors, storage, or GPS frequency are typically used. In this paper, we propose a VPR mapping approach that uses two reference traverses from the operating environment to automatically select an appropriate map density satisfying two user-defined requirements: (1) a target Local Recall@1 level, and (2) the proportion of the operational environment over which it must be met or exceeded, which we term the Recall Achievement Rate (RAR). Our approach uses spatial consistency and coherence features from reference-to-reference matches at multiple map densities to estimate the density needed to meet these targets on unseen deployment data. Through experiments across two VPR methods and the Nordland and Oxford RobotCar benchmarks, we show that our system consistently meets or exceeds the target Local Recall@1 over at least the user-specified proportion of the environment. Comparisons with alternative baselines show that it reliably selects an appropriate operating point in map density, avoiding unnecessarily dense maps. Finally, ablations evaluate sensitivity to reference traversal choice and segment length, and our analysis reveals that conventional global Recall@1 is a poor predictor of the often more operationally meaningful RAR metric. --- ### The Design Space of Tri-Modal Masked Diffusion Models Authors: Louis Bethune, Victor Turrisi, Bruno Kacper Mlodozeniec, Pau Rodriguez Lopez, Lokesh Boominathan, Nikhil Bhendawade, Amitis Shidani, Joris Pelemans, Theo X. Olausson, Devon Hjelm, Paul Dixon, Joao Monteiro et al. Published: 2026-02-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.21472v1 Discrete diffusion models have emerged as strong alternatives to autoregressive language models, with recent work initializing and fine-tuning a base unimodal model for bimodal generation. Diverging from previous approaches, we introduce the first tri-modal masked diffusion model pretrained from scratch on text, image-text, and audio-text data. We systematically analyze multimodal scaling laws, modality mixing ratios, noise schedules, and batch-size effects, and we provide optimized inference sampling defaults. Our batch-size analysis yields a novel stochastic differential equation (SDE)-based reparameterization that eliminates the need for tuning the optimal batch size as reported in recent work. This reparameterization decouples the physical batch size, often chosen based on compute constraints (GPU saturation, FLOP efficiency, wall-clock time), from the logical batch size, chosen to balance gradient variance during stochastic optimization. Finally, we pretrain a preliminary 3B-parameter tri-modal model on 6.4T tokens, demonstrating the capabilities of a unified design and achieving strong results in text generation, text-to-image tasks, and text-to-speech tasks. Our work represents the largest-scale systematic open study of multimodal discrete diffusion models conducted to date, providing insights into scaling behaviors across multiple modalities. --- ### D-Flow SGLD: Source-Space Posterior Sampling for Scientific Inverse Problems with Flow Matching Authors: Meet Hemant Parikh, Yaqin Chen, Jian-Xun Wang Published: 2026-02-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.21469v1 Data assimilation and scientific inverse problems require reconstructing high-dimensional physical states from sparse and noisy observations, ideally with uncertainty-aware posterior samples that remain faithful to learned priors and governing physics. While training-free conditional generation is well developed for diffusion models, corresponding conditioning and posterior sampling strategies for Flow Matching (FM) priors remain comparatively under-explored, especially on scientific benchmarks where fidelity must be assessed beyond measurement misfit. In this work, we study training-free conditional generation for scientific inverse problems under FM priors and organize existing inference-time strategies by where measurement information is injected: (i) guided transport dynamics that perturb sampling trajectories using likelihood information, and (ii) source-distribution inference that performs posterior inference over the source variable while keeping the learned transport fixed. Building on the latter, we propose D-Flow SGLD, a source-space posterior sampling method that augments differentiable source inference with preconditioned stochastic gradient Langevin dynamics, enabling scalable exploration of the source posterior induced by new measurement operators without retraining the prior or modifying the learned FM dynamics. We benchmark representative methods from both families on a hierarchy of problems: 2D toy posteriors, chaotic Kuramoto-Sivashinsky trajectories, and wall-bounded turbulence reconstruction. Across these settings, we quantify trade-offs among measurement assimilation, posterior diversity, and physics/statistics fidelity, and establish D-Flow SGLD as a practical FM-compatible posterior sampler for scientific inverse problems. --- ### Unsupervised Discovery of Intermediate Phase Order in the Frustrated $J_1$-$J_2$ Heisenberg Model via Prometheus Framework Authors: Brandon Yee, Wilson Collins, Maximilian Rutkowski Published: 2026-02-25 Categories: cond-mat.str-el, cond-mat.dis-nn, cs.LG, quant-ph arXiv: https://arxiv.org/abs/2602.21468v5 The spin-$1/2$ $J_1$-$J_2$ Heisenberg model on the square lattice exhibits a debated intermediate phase between Néel antiferromagnetic and stripe ordered regimes, with competing theories proposing plaquette valence bond, nematic, and quantum spin liquid ground states. We apply the Prometheus variational autoencoder framework -- previously applied to classical (2D, 3D Ising) and quantum (disordered transverse field Ising) phase transitions -- to systematically explore the $J_1$-$J_2$ phase diagram using a multi-scale approach. For $L=4$, we employ exact diagonalization with full wavefunction analysis via quantum-aware VAE. For larger systems ($L=6, 8$), we introduce a reduced density matrix (RDM) based methodology using DMRG ground states, enabling scaling beyond the exponential barrier of full Hilbert space representation. Through dense parameter scans of $J_2/J_1 \in (0, 1)$ and comprehensive latent space analysis, we identify the structure factor $S(π,π)$ and $S(π,0)$ as the dominant order parameters discovered by the VAE, with correlations exceeding $|r| > 0.97$. The RDM-VAE approach successfully captures the Néel-to-stripe crossover near $J_2/J_1 \approx 0.5$--$0.6$, demonstrating that local quantum correlations encoded in reduced density matrices contain sufficient information for unsupervised phase discovery. This work establishes a scalable pathway for applying machine learning to frustrated quantum systems where full wavefunction access is computationally prohibitive. --- ### Geometric Priors for Generalizable World Models via Vector Symbolic Architecture Authors: William Youngwoo Chung, Calvin Yeung, Hansen Jin Lillemark, Zhuowen Zou, Xiangjian Liu, Mohsen Imani Published: 2026-02-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.21467v1 A key challenge in artificial intelligence and neuroscience is understanding how neural systems learn representations that capture the underlying dynamics of the world. Most world models represent the transition function with unstructured neural networks, limiting interpretability, sample efficiency, and generalization to unseen states or action compositions. We address these issues with a generalizable world model grounded in Vector Symbolic Architecture (VSA) principles as geometric priors. Our approach utilizes learnable Fourier Holographic Reduced Representation (FHRR) encoders to map states and actions into a high dimensional complex vector space with learned group structure and models transitions with element-wise complex multiplication. We formalize the framework's group theoretic foundation and show how training such structured representations to be approximately invariant enables strong multi-step composition directly in latent space and generalization performances over various experiments. On a discrete grid world environment, our model achieves 87.5% zero shot accuracy to unseen state-action pairs, obtains 53.6% higher accuracy on 20-timestep horizon rollouts, and demonstrates 4x higher robustness to noise relative to an MLP baseline. These results highlight how training to have latent group structure yields generalizable, data-efficient, and interpretable world models, providing a principled pathway toward structured models for real-world planning and reasoning. --- ### Asymptotically Fast Clebsch-Gordan Tensor Products with Vector Spherical Harmonics Authors: YuQing Xie, Ameya Daigavane, Mit Kotak, Tess Smidt Published: 2026-02-25 Categories: cs.LG, physics.comp-ph arXiv: https://arxiv.org/abs/2602.21466v1 $E(3)$-equivariant neural networks have proven to be effective in a wide range of 3D modeling tasks. A fundamental operation of such networks is the tensor product, which allows interaction between different feature types. Because this operation scales poorly, there has been considerable work towards accelerating this interaction. However, recently \citet{xieprice} have pointed out that most speedups come from a reduction in expressivity rather than true algorithmic improvements on computing Clebsch-Gordan tensor products. A modification of Gaunt tensor product \citep{gaunt} can give a true asymptotic speedup but is incomplete and misses many interactions. In this work, we provide the first complete algorithm which truly provides asymptotic benefits Clebsch-Gordan tensor products. For full CGTP, our algorithm brings runtime complexity from the naive $O(L^6)$ to $O(L^4\log^2 L)$, close to the lower bound of $O(L^4)$. We first show how generalizing fast Fourier based convolution naturally leads to the previously proposed Gaunt tensor product \citep{gaunt}. To remedy antisymmetry issues, we generalize from scalar signals to irrep valued signals, giving us tensor spherical harmonics. We prove a generalized Gaunt formula for the tensor harmonics. Finally, we show that we only need up to vector valued signals to recover the missing interactions of Gaunt tensor product. --- ### iMiGUE-Speech: A Spontaneous Speech Dataset for Affective Analysis Authors: Sofoklis Kakouros, Fang Kang, Haoyu Chen Published: 2026-02-25 Categories: eess.AS, cs.CL arXiv: https://arxiv.org/abs/2602.21464v1 This work presents iMiGUE-Speech, an extension of the iMiGUE dataset that provides a spontaneous affective corpus for studying emotional and affective states. The new release focuses on speech and enriches the original dataset with additional metadata, including speech transcripts, speaker-role separation between interviewer and interviewee, and word-level forced alignments. Unlike existing emotional speech datasets that rely on acted or laboratory-elicited emotions, iMiGUE-Speech captures spontaneous affect arising naturally from real match outcomes. To demonstrate the utility of the dataset and establish initial benchmarks, we introduce two evaluation tasks for comparative assessment: speech emotion recognition and transcript-based sentiment analysis. These tasks leverage state-of-the-art pre-trained representations to assess the dataset's ability to capture spontaneous affective states from both acoustic and linguistic modalities. iMiGUE-Speech can also be synchronously paired with micro-gesture annotations from the original iMiGUE dataset, forming a uniquely multimodal resource for studying speech-gesture affective dynamics. The extended dataset is available at https://github.com/CV-AC/imigue-speech. --- ### Effects of Training Data Quality on Classifier Performance Authors: Alan F. Karr, Regina Ruane Published: 2026-02-25 Categories: cs.LG, q-bio.GN, stat.ML arXiv: https://arxiv.org/abs/2602.21462v1 We describe extensive numerical experiments assessing and quantifying how classifier performance depends on the quality of the training data, a frequently neglected component of the analysis of classifiers. More specifically, in the scientific context of metagenomic assembly of short DNA reads into "contigs," we examine the effects of degrading the quality of the training data by multiple mechanisms, and for four classifiers -- Bayes classifiers, neural nets, partition models and random forests. We investigate both individual behavior and congruence among the classifiers. We find breakdown-like behavior that holds for all four classifiers, as degradation increases and they move from being mostly correct to only coincidentally correct, because they are wrong in the same way. In the process, a picture of spatial heterogeneity emerges: as the training data move farther from analysis data, classifier decisions degenerate, the boundary becomes less dense, and congruence increases. --- ### VecGlypher: Unified Vector Glyph Generation with Language Models Authors: Xiaoke Huang, Bhavul Gauri, Kam Woh Ng, Tony Ng, Mengmeng Xu, Zhiheng Liu, Weiming Ren, Zhaochong An, Zijian Zhou, Haonan Qiu, Yuyin Zhou, Sen He et al. Published: 2026-02-25 Categories: cs.CL arXiv: https://arxiv.org/abs/2602.21461v1 Vector glyphs are the atomic units of digital typography, yet most learning-based pipelines still depend on carefully curated exemplar sheets and raster-to-vector postprocessing, which limits accessibility and editability. We introduce VecGlypher, a single multimodal language model that generates high-fidelity vector glyphs directly from text descriptions or image exemplars. Given a style prompt, optional reference glyph images, and a target character, VecGlypher autoregressively emits SVG path tokens, avoiding raster intermediates and producing editable, watertight outlines in one pass. A typography-aware data and training recipe makes this possible: (i) a large-scale continuation stage on 39K noisy Envato fonts to master SVG syntax and long-horizon geometry, followed by (ii) post-training on 2.5K expert-annotated Google Fonts with descriptive tags and exemplars to align language and imagery with geometry; preprocessing normalizes coordinate frames, canonicalizes paths, de-duplicates families, and quantizes coordinates for stable long-sequence decoding. On cross-family OOD evaluation, VecGlypher substantially outperforms both general-purpose LLMs and specialized vector-font baselines for text-only generation, while image-referenced generation reaches a state-of-the-art performance, with marked gains over DeepVecFont-v2 and DualVector. Ablations show that model scale and the two-stage recipe are critical and that absolute-coordinate serialization yields the best geometry. VecGlypher lowers the barrier to font creation by letting users design with words or exemplars, and provides a scalable foundation for future multimodal design tools. --- ### Revisiting Text Ranking in Deep Research Authors: Chuan Meng, Litu Ou, Sean MacAvaney, Jeff Dalton Published: 2026-02-25 Categories: cs.IR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2602.21456v2 Deep research has emerged as an important task that aims to address hard queries that need extensive open-web exploration. To tackle it, most prior work equips large language model (LLM)-based agents with opaque web search APIs, enabling agents to iteratively issue search queries, retrieve external evidence, and reason over it. Despite search's essential role in deep research, black-box web search APIs leave the behaviour of established text ranking methods in deep research largely unclear. To fill this gap, we reproduce key findings and best practices for text ranking methods in deep research. We examine their effectiveness from three perspectives: (i) retrieval units (documents vs. passages), (ii) pipeline configurations (different retrievers, re-rankers, and re-ranking depths), and (iii) query characteristics (the mismatch between agent-issued queries and the training queries of text rankers). We perform experiments on BrowseComp-Plus, a deep research dataset with a fixed corpus, evaluating 2 open-source agents, 5 retrievers, and 3 re-rankers. We find that agent-issued queries typically follow web-search-style syntax (e.g., quoted exact matches), favouring lexical, learned sparse, and multi-vector retrievers; passage-level units are more efficient under limited context windows, and avoid the difficulties of document length normalisation in lexical retrieval; re-ranking is highly effective. We further propose a query-to-question (Q2Q) method that translates agent-issued queries into natural language questions, significantly reducing the query mismatch. --- ### When Learning Hurts: Fixed-Pole RNN for Real-Time Online Training Authors: Alexander Morgan, Ummay Sumaya Khan, Lingjia Liu, Lizhong Zheng Published: 2026-02-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2602.21454v1 Recurrent neural networks (RNNs) can be interpreted as discrete-time state-space models, where the state evolution corresponds to an infinite-impulse-response (IIR) filtering operation governed by both feedforward weights and recurrent poles. While, in principle, all parameters including pole locations can be optimized via backpropagation through time (BPTT), such joint learning incurs substantial computational overhead and is often impractical for applications with limited training data. Echo state networks (ESNs) mitigate this limitation by fixing the recurrent dynamics and training only a linear readout, enabling efficient and stable online adaptation. In this work, we analytically and empirically examine why learning recurrent poles does not provide tangible benefits in data-constrained, real-time learning scenarios. Our analysis shows that pole learning renders the weight optimization problem highly non-convex, requiring significantly more training samples and iterations for gradient-based methods to converge to meaningful solutions. Empirically, we observe that for complex-valued data, gradient descent frequently exhibits prolonged plateaus, and advanced optimizers offer limited improvement. In contrast, fixed-pole architectures induce stable and well-conditioned state representations even with limited training data. Numerical results demonstrate that fixed-pole networks achieve superior performance with lower training complexity, making them more suitable for online real-time tasks. --- ### Adversarial Robustness of Deep Learning-Based Thyroid Nodule Segmentation in Ultrasound Authors: Nicholas Dietrich, David McShannon Published: 2026-02-25 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2602.21452v1 Introduction: Deep learning-based segmentation models are increasingly integrated into clinical imaging workflows, yet their robustness to adversarial perturbations remains incompletely characterized, particularly for ultrasound images. We evaluated adversarial attacks and inference-time defenses for thyroid nodule segmentation in B-mode ultrasound. Methods: Two black-box adversarial attacks were developed: (1) Structured Speckle Amplification Attack (SSAA), which injects boundary-targeted noise, and (2) Frequency-Domain Ultrasound Attack (FDUA), which applies bandpass-filtered phase perturbations in the Fourier domain. Three inference-time mitigations were evaluated on adversarial images: randomized preprocessing with test-time augmentation, deterministic input denoising, and stochastic ensemble inference with consistency-aware aggregation. Experiments were conducted on a U-Net segmentation model trained on cine-clips from a database of 192 thyroid nodules. Results: The baseline model achieved a mean Dice similarity coefficient (DSC) of 0.76 (SD 0.20) on unperturbed images. SSAA reduced DSC by 0.29 (SD 0.20) while maintaining high visual similarity (SSIM = 0.94). FDUA resulted in a smaller DSC reduction of 0.11 (SD 0.09) with lower visual fidelity (SSIM = 0.82). Against SSAA, all three defenses significantly improved DSC after correction, with deterministic denoising showing the largest recovery (+0.10, p < 0.001), followed by randomized preprocessing (+0.09, p < 0.001), and stochastic ensemble inference (+0.08, p = 0.002). No defense achieved statistically significant improvement against FDUA. Conclusion: Spatial-domain adversarial perturbations in ultrasound segmentation showed partial mitigation with input preprocessing, whereas frequency-domain perturbations were not mitigated by the defenses, highlighting modality-specific challenges in adversarial robustness evaluation. --- ### Mirror Descent-Type Algorithms for the Variational Inequality Problem with Functional Constraints Authors: Mohammad S. Alkousa, Fedor S. Stonyakin, Belal A. Alashqar, Seydamet S. Ablaev Published: 2026-02-24 Categories: cs.LG, math.OC arXiv: https://arxiv.org/abs/2605.16262v1 Variational inequalities play a key role in machine learning research, such as generative adversarial networks, reinforcement learning, adversarial training, and generative models. This paper is devoted to the constrained variational inequality problems with functional constraints (inequality-type constraints). We propose some mirror descent-type algorithms that switch between productive and non-productive steps depending on the values of the functional constraints at iterations, with many different step size rules and stopping criteria. We analyze the proposed algorithms and prove their optimal convergence rate to achieve a solution with desired accuracy, for problems with bounded and monotone operators and Lipschitz convex functional constraints. In addition, we propose a modification of the proposed algorithms by considering each functional constraint in the calculation when we have a productive step, as well as the first constraint that violates the feasibility. This modification can save the running time of algorithms when we have many functional constraints. In addition, we provide an analysis of the proposed algorithms for $δ$-monotone operators, allowing us to apply the proposed algorithms, as a special case, to constrained minimization problems when we do not have access to the exact information about the subgradient of the objective function. Numerical experiments that illustrate the work and performance of the proposed algorithms are also given. --- ### ERA: Evidence-based Reliability Alignment for Honest Retrieval-Augmented Generation Authors: Sunguk Shin, Meeyoung Cha, Byung-Jun Lee, Sungwon Park Published: 2026-02-24 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2604.20854v1 Retrieval-Augmented Generation (RAG) grounds language models in factual evidence but introduces critical challenges regarding knowledge conflicts between internalized parameters and retrieved information. However, existing reliability methods, typically relying on scalar confidence, fail to explicitly distinguish between epistemic uncertainty and inherent data ambiguity in such hybrid scenarios. In this paper, we propose a new framework called ERA (Evidence-based Reliability Alignment) to enhance abstention behavior in RAG systems by shifting confidence estimation from scalar probabilities to explicit evidence distributions. Our method consists of two main components: (1) Contextual Evidence Quantification, which models internal and external knowledge as independent belief masses via the Dirichlet distribution, and (2) Quantifying Knowledge Conflict, which leverages Dempster-Shafer Theory (DST) to rigorously measure the geometric discordance between information sources. These components are used to disentangle epistemic uncertainty from aleatoric uncertainty and modulate the optimization objective based on detected conflicts. Experiments on standard benchmarks and a curated generalization dataset demonstrate that our approach significantly outperforms baselines, optimizing the trade-off between answer coverage and abstention with superior calibration. --- ### How Robustly do LLMs Understand Execution Semantics? Authors: Claudio Spiess, Prem Devanbu, Earl T. Barr Published: 2026-02-24 Categories: cs.SE, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.16320v1 LLMs demonstrate remarkable reasoning capabilities, yet whether they utilize internal world models or rely on sophisticated pattern matching remains open. We study LLMs through the lens of robustness of their code understanding using a standard program-output prediction task. Our results reveal a stark divergence in model behavior: while open-source reasoning models (DeepSeek-R1 family) maintain stable, albeit somewhat lower accuracies (38% to 67%) under code transformations & input perturbations, the frontier model GPT-5.2 exhibits significant brittleness. Despite achieving a near-perfect score of 99% on the original, unperturbed CRUXEval benchmark, perturbed inputs trigger accuracy declines between 20% and 24%. In addition, we find that many models perform much worse at predicting behavior on perturbed inputs that raise exceptions, and that prediction performance depends on the kind of exception. We study remedies to address this deficiency in exception prediction, and evaluate the effect of these remedies on the ability to predict non-exception behaviors. Our findings both point to limitations in the way all models understand code, and establish the value of using perturbation to evaluate code models. --- ### Explainable Planning for Hybrid Systems Authors: Mir Md Sajid Sarwar Published: 2026-02-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.09578v1 The recent advancement in artificial intelligence (AI) technologies facilitates a paradigm shift toward automation. Autonomous systems are fully or partially replacing manually crafted ones. At the core of these systems is automated planning. With the advent of powerful planners, automated planning is now applied to many complex and safety-critical domains, including smart energy grids, self-driving cars, warehouse automation, urban and air traffic control, search and rescue operations, surveillance, robotics, and healthcare. There is a growing need to generate explanations of AI-based systems, which is one of the major challenges the planning community faces today. The thesis presents a comprehensive study on explainable artificial intelligence planning (XAIP) for hybrid systems that capture a representation of real-world problems closely. --- ### Generative UI: LLMs are Effective UI Generators Authors: Yaniv Leviathan, Dani Valevski, Matan Kalman, Danny Lumen, Eyal Segalis, Eyal Molad, Shlomi Pasternak, Vishnu Natchu, Valerie Nygaard, Srinivasan, Venkatachary, James Manyika et al. Published: 2026-02-24 Categories: cs.HC, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.09577v1 AI models excel at creating content, but typically render it with static, predefined interfaces. Specifically, the output of LLMs is often a markdown "wall of text". Generative UI is a long standing promise, where the model generates not just the content, but the interface itself. Until now, Generative UI was not possible in a robust fashion. We demonstrate that when properly prompted and equipped with the right set of tools, a modern LLM can robustly produce high quality custom UIs for virtually any prompt. When ignoring generation speed, results generated by our implementation are overwhelmingly preferred by humans over the standard LLM markdown output. In fact, while the results generated by our implementation are worse than those crafted by human experts, they are at least comparable in 50% of cases. We show that this ability for robust Generative UI is emergent, with substantial improvements from previous models. We also create and release PAGEN, a novel dataset of expert-crafted results to aid in evaluating Generative UI implementations, as well as the results of our system for future comparisons. Interactive examples can be seen at https://generativeui.github.io --- ### AHC: Meta-Learned Adaptive Compression for Continual Object Detection on Memory-Constrained Microcontrollers Authors: Bibin Wilson Published: 2026-02-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.09576v1 Deploying continual object detection on microcontrollers (MCUs) with under 100KB memory requires efficient feature compression that can adapt to evolving task distributions. Existing approaches rely on fixed compression strategies (e.g., FiLM conditioning) that cannot adapt to heterogeneous task characteristics, leading to suboptimal memory utilization and catastrophic forgetting. We introduce Adaptive Hierarchical Compression (AHC), a meta-learning framework featuring three key innovations: (1) true MAML-based compression that adapts via gradient descent to each new task in just 5 inner-loop steps, (2) hierarchical multi-scale compression with scale-aware ratios (8:1 for P3, 6.4:1 for P4, 4:1 for P5) matching FPN redundancy patterns, and (3) a dual-memory architecture combining short-term and long-term banks with importance-based consolidation under a hard 100KB budget. We provide formal theoretical guarantees bounding catastrophic forgetting as O(ε{sq.root(T)} + 1/{sq.root(M)}) where ε is compression error, T is task count, and M is memory size. Experiments on CORe50, TiROD, and PASCAL VOC benchmarks with three standard baselines (Fine-tuning,EWC, iCaRL) demonstrate that AHC enables practical continual detection within a 100KB replay budget, achieving competitive accuracy through mean-pooled compressed feature replay combined with EWC regularization and feature distillation. --- ### Turing Test on Screen: A Benchmark for Mobile GUI Agent Humanization Authors: Jiachen Zhu, Lingyu Yang, Rong Shan, Congmin Zheng, Zeyu Zheng, Weiwen Liu, Yong Yu, Weinan Zhang, Jianghao Lin Published: 2026-02-24 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.09574v1 The rise of autonomous GUI agents has triggered adversarial countermeasures from digital platforms, yet existing research prioritizes utility and robustness over the critical dimension of anti-detection. We argue that for agents to survive in human-centric ecosystems, they must evolve Humanization capabilities. We introduce the ``Turing Test on Screen,'' formally modeling the interaction as a MinMax optimization problem between a detector and an agent aiming to minimize behavioral divergence. We then collect a new high-fidelity dataset of mobile touch dynamics, and conduct our analysis that vanilla LMM-based agents are easily detectable due to unnatural kinematics. Consequently, we establish the Agent Humanization Benchmark (AHB) and detection metrics to quantify the trade-off between imitability and utility. Finally, we propose methods ranging from heuristic noise to data-driven behavioral matching, demonstrating that agents can achieve high imitability theoretically and empirically without sacrificing performance. This work shifts the paradigm from whether an agent can perform a task to how it performs it within a human-centric ecosystem, laying the groundwork for seamless coexistence in adversarial digital environments. --- ### High-Precision Estimation of the State-Space Complexity of Shogi via the Monte Carlo Method Authors: Sotaro Ishii, Tetsuro Tanaka Published: 2026-02-24 Categories: cs.AI, cs.GT arXiv: https://arxiv.org/abs/2604.06189v1 Determining the state-space complexity of the game of Shogi (Japanese Chess) has been a challenging problem, with previous combinatorial estimates leaving a gap of five orders of magnitude ($10^{64}$ to $10^{69}$). This large gap arises from the difficulty of distinguishing Shogi positions legally reachable from the initial position among the vast number of valid board configurations. In this paper, we present a high-precision statistical estimation of the number of reachable positions in Shogi. Our method combines Monte Carlo sampling with a novel reachability test that utilizes a reverse search toward a set of "King-King only" (KK) positions, rather than a single-target backward search to the single initial position. This approach significantly reduces the search effort for determining unreachability. Based on a sample of 5 billion positions, we estimated the number of legal positions in Shogi to be $6.55 \times 10^{68}$ (to three significant digits) with a $3σ$ confidence level, substantially improving upon previously known bounds. We also applied this method to Mini Shogi, determining its complexity to be approximately $2.38 \times 10^{18}$. --- ### Can AI be a Teaching Partner? Evaluating ChatGPT, Gemini, and DeepSeek across Three Teaching Strategies Authors: Talita de Paula Cypriano de Souza, Shruti Mehta, Matheus Arataque Uema, Luciano Bernardes de Paula, Seiji Isotani Published: 2026-02-24 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2603.26673v1 There are growing promises that Large Language Models (LLMs) can support students' learning by providing explanations, feedback, and guidance. However, despite their rapid adoption and widespread attention, there is still limited empirical evidence regarding the pedagogical skills of LLMs. This article presents a comparative study of popular LLMs, namely, ChatGPT, DeepSeek, and Gemini, acting as teaching agents. An evaluation protocol was developed, focusing on three pedagogical strategies: Examples, Explanations and Analogies, and the Socratic Method. Six human judges conducted the evaluations in the context of teaching the C programming language to beginners. The results indicate that LLM models exhibited similar interaction patterns in the pedagogical strategies of Examples and Explanations and Analogies. In contrast, for the Socratic Method, the models showed greater sensitivity to the pedagogical strategy and the initial prompt. Overall, ChatGPT and Gemini received higher scores, whereas DeepSeek obtained lower scores across the criteria, indicating differences in pedagogical performance across models. --- ### Frayed RoPE and Long Inputs: A Geometric Perspective Authors: Davis Wertheimer, Aozhong Zhang, Derrick Liu, Penghang Yin, Naigang Wang Published: 2026-02-24 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2603.18017v1 Rotary Positional Embedding (RoPE) is a widely adopted technique for encoding position in language models, which, while effective, causes performance breakdown when input length exceeds training length. Prior analyses assert (rightly) that long inputs cause channels to rotate ``out of distribution,'' but it is not clear how extra rotation relates to or causes pathological behavior. Through empirical and theoretical analysis we advance a unified geometric understanding of attention behavior with RoPE. We find that attention induces tight clustering of separated key and query latent point clouds, allowing for creation of sink tokens: placeholders that allow attention heads to avoid token mixing when not required. RoPE applied to longer inputs damages this key/query cluster separation, producing pathological behavior by inhibiting sink token functionality. From this geometric perspective, we propose RoPE-ID (In Distribution), a straightforward modification that allows attention layers to generalize to longer inputs out of the box: apply RoPE with high frequency to a subset of channels. We demonstrate the effectiveness of RoPE-ID for extended inputs using 1B and 3B parameter Transformers on the LongBench and RULER information retrieval benchmarks. --- ### MineDraft: A Framework for Batch Parallel Speculative Decoding Authors: Zhenwei Tang, Arun Verma, Zijian Zhou, Zhaoxuan Wu, Alok Prakash, Daniela Rus, Bryan Kian Hsiang Low Published: 2026-02-24 Categories: cs.CL, cs.AI, cs.DC, cs.LG arXiv: https://arxiv.org/abs/2603.18016v3 Speculative decoding (SD) accelerates large language model inference by using a smaller draft model to propose draft tokens that are subsequently verified by a larger target model. However, the performance of standard SD is often limited by the strictly sequential execution of these drafting and verification stages. To address this, this paper proposes MineDraft, a batch parallel speculative decoding (PSD) framework designed to effectively hide drafting latency by overlapping it with verification. Our theoretical analysis shows that PSD is substantially more efficient than standard SD. MineDraft realizes the PSD through a novel batch-parallel design that maintains two batches of requests, overlapping drafting for one batch with verification for the other. Our experimental results show significant improvements of \alg{} in both throughput (up to 75%) and end-to-end latency (up to 39%) over standard SD. Furthermore, we have implemented MineDraft as a plugin for vLLM, demonstrating its practicality for production-ready inference systems. --- ### MineDraft: A Framework for Batch Parallel Speculative Decoding Authors: Zhenwei Tang, Arun Verma, Zijian Zhou, Zhaoxuan Wu, Alok Prakash, Daniela Rus, Bryan Kian Hsiang Low Published: 2026-02-24 Categories: cs.CL, cs.AI, cs.DC, cs.LG arXiv: https://arxiv.org/abs/2603.18016v2 Speculative decoding (SD) accelerates large language model inference by using a smaller draft model to propose draft tokens that are subsequently verified by a larger target model. However, the performance of standard SD is often limited by the strictly sequential execution of these drafting and verification stages. To address this, this paper proposes MineDraft, a batch parallel speculative decoding (PSD) framework designed to effectively hide drafting latency by overlapping it with verification. Our theoretical analysis shows that PSD is substantially more efficient than standard SD. MineDraft realizes the PSD through a novel batch-parallel design that maintains two batches of requests, overlapping drafting for one batch with verification for the other. Our experimental results show significant improvements of MineDraft in both throughput (up to 75%) and end-to-end latency (up to 39%) over standard SD. Furthermore, we have implemented MineDraft as a plugin for vLLM, demonstrating its practicality for production-ready inference systems. --- ### Beyond Accuracy: An Explainability-Driven Analysis of Harmful Content Detection Authors: Trishita Dhara, Siddhesh Sheth Published: 2026-02-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.18015v1 Although automated harmful content detection systems are frequently used to monitor online platforms, moderators and end users frequently cannot understand the logic underlying their predictions. While recent studies have focused on increasing classification accuracy, little focus has been placed on comprehending why neural models identify content as harmful, especially when it comes to borderline, contextual, and politically sensitive situations. In this work, a neural harmful content detection model trained on the Civil Comments dataset is analyzed explainability-drivenly. Two popular post-hoc explanation methods, Shapley Additive Explanations and Integrated Gradients, are used to analyze the behavior of a RoBERTa-based classifier in both correct predictions and systematic failure cases. Despite strong overall performance, with an area under the curve of 0.93 and an accuracy of 0.94, the analysis reveals limitations that are not observable from aggregate evaluation metrics alone. Integrated Gradients appear to extract more diffuse contextual attributions while Shapley Additive Explanations extract more focused attributions on explicit lexical cues. The consequent divergence in their outputs manifests in both false negatives and false positives. Qualitative case studies reveal recurring failure modes such as indirect toxicity, lexical over-attribution, or political discourse. The results suggest that explainable AI can foster human-in-the-loop moderation by exposing model uncertainty and increasing the interpretable rationale behind automated decisions. Most importantly, this work highlights the role of explainability as a transparency and diagnostic resource for online harmful content detection systems rather than as a performance-enhancing lever. --- ### Real-Time Trustworthiness Scoring for LLM Structured Outputs and Data Extraction Authors: Hui Wen Goh, Jonas Mueller Published: 2026-02-24 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2603.18014v2 Structured Outputs from current LLMs exhibit sporadic errors, hindering enterprise AI deployment. We present CONSTRUCT, a real-time uncertainty estimator that scores the trustworthiness of LLM Structured Outputs. Lower-scoring outputs are more likely to contain errors, enabling automatic prioritization of limited human review bandwidth. CONSTRUCT additionally scores the trustworthiness of each field within a Structured Output, helping reviewers quickly identify which parts of the output are incorrect. Our method is suitable for any LLM (including black-box LLM APIs without logprobs), does not require labeled training data or custom model deployment, and supports complex Structured Outputs with heterogeneous fields and nested JSON schemas. We also introduce one of the first public LLM Structured Output benchmarks with reliable ground-truth values. Over this four-dataset benchmark, CONSTRUCT detects errors in outputs from various LLMs (including Gemini 3 and GPT-5) with significantly higher precision/recall than existing techniques. --- ### Learned but Not Expressed: Capability-Expression Dissociation in Large Language Models Authors: Toshiyuki Shigemura Published: 2026-02-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.18013v1 Large language models (LLMs) demonstrate the capacity to reconstruct and trace learned content from their training data under specific elicitation conditions, yet this capability does not manifest in standard generation contexts. This empirical observational study examines the expression of non-causal, non-implementable solution types across 300 prompt-response generations spanning narrative and problem-solving task contexts. Drawing on recent findings regarding memorization contiguity and alignment-induced discourse priors, we document a systematic dissociation between learned capability and expressed output. Across three distinct LLMs, ten task scenarios, and both creative narrative and practical advisory contexts, we documented zero instances of non-causal solution frames in generated outputs (0%, 95% CI: (0%, 1.2%)), despite verified reconstruction capability under conditional extraction. These findings challenge the prevailing assumption that training data presence directly predicts output probability, demonstrating instead that task-conditioned generation policies can comprehensively suppress learned content across diverse contexts. The results offer implications for understanding generation dynamics, output distribution control, and the behavioral boundaries of contemporary LLMs. --- ### DynaRAG: Bridging Static and Dynamic Knowledge in Retrieval-Augmented Generation Authors: Penghao Liang, Mengwei Yuan, Jianan Liu, Jing Yang, Xianyou Li, Weiran Yan, Yichao Wu Published: 2026-02-24 Categories: cs.CL, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2603.18012v1 We present DynaRAG, a retrieval-augmented generation (RAG) framework designed to handle both static and time-sensitive information needs through dynamic knowledge integration. Unlike traditional RAG pipelines that rely solely on static corpora, DynaRAG selectively invokes external APIs when retrieved documents are insufficient for answering a query. The system employs an LLM-based reranker to assess document relevance, a sufficiency classifier to determine when fallback is necessary, and Gorilla v2 -- a state-of-the-art API calling model -- for accurate tool invocation. We further enhance robustness by incorporating schema filtering via FAISS to guide API selection. Evaluations on the CRAG benchmark demonstrate that DynaRAG significantly improves accuracy on dynamic questions, while also reducing hallucinations. Our results highlight the importance of dynamic-aware routing and selective tool use in building reliable, real-world question-answering systems. --- ### Federated Multi Agent Deep Learning and Neural Networks for Advanced Distributed Sensing in Wireless Networks Authors: Nadine Muller, Stefano DeRosa, Su Zhang, Chun Lee Huan Published: 2026-02-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.16881v1 Multi-agent deep learning (MADL), including multi-agent deep reinforcement learning (MADRL), distributed/federated training, and graph-structured neural networks, is becoming a unifying framework for decision-making and inference in wireless systems where sensing, communication, and computing are tightly coupled. Recent 5G-Advanced and 6G visions strengthen this coupling through integrated sensing and communication, edge intelligence, open programmable RAN, and non-terrestrial/UAV networking, which create decentralized, partially observed, time-varying, and resource-constrained control problems. This survey synthesizes the state of the art, with emphasis on 2021-2025 research, on MADL for distributed sensing and wireless communications. We present a task-driven taxonomy across (i) learning formulations (Markov games, Dec-POMDPs, CTDE), (ii) neural architectures (GNN-based radio resource management, attention-based policies, hierarchical learning, and over-the-air aggregation), (iii) advanced techniques (federated reinforcement learning, communication-efficient federated deep RL, and serverless edge learning orchestration), and (iv) application domains (MEC offloading with slicing, UAV-enabled heterogeneous networks with power-domain NOMA, intrusion detection in sensor networks, and ISAC-driven perceptive mobile networks). We also provide comparative tables of algorithms, training topologies, and system-level trade-offs in latency, spectral efficiency, energy, privacy, and robustness. Finally, we identify open issues including scalability, non-stationarity, security against poisoning and backdoors, communication overhead, and real-time safety, and outline research directions toward 6G-native sense-communicate-compute-learn systems. --- ### NeuroNarrator: A Generalist EEG-to-Text Foundation Model for Clinical Interpretation via Spectro-Spatial Grounding and Temporal State-Space Reasoning Authors: Guoan Wang, Shihao Yang, Jun-en Ding, Hao Zhu, Feng Liu Published: 2026-02-24 Categories: eess.SP, cs.CL, cs.LG, q-bio.NC arXiv: https://arxiv.org/abs/2603.16880v2 Electroencephalography (EEG) provides a non-invasive window into neural dynamics at high temporal resolution and plays a pivotal role in clinical neuroscience research. Despite this potential, prevailing computational approaches to EEG analysis remain largely confined to task-specific classification objectives or coarse-grained pattern recognition, offering limited support for clinically meaningful interpretation. To address these limitations, we introduce NeuroNarrator, the first generalist EEG-to-text foundation model designed to translate electrophysiological segments into precise clinical narratives. A cornerstone of this framework is the curation of NeuroCorpus-160K, the first harmonized large-scale resource pairing over 160,000 EEG segments with structured, clinically grounded natural-language descriptions. Our architecture first aligns temporal EEG waveforms with spatial topographic maps via a rigorous contrastive objective, establishing spectro-spatially grounded representations. Building on this grounding, we condition a Large Language Model through a state-space-inspired formulation that integrates historical temporal and spectral context to support coherent clinical narrative generation. This approach establishes a principled bridge between continuous signal dynamics and discrete clinical language, enabling interpretable narrative generation that facilitates expert interpretation and supports clinical reporting workflows. Extensive evaluations across diverse benchmarks and zero-shot transfer tasks highlight NeuroNarrator's capacity to integrate temporal, spectral, and spatial dynamics, positioning it as a foundational framework for time-frequency-aware, open-ended clinical interpretation of electrophysiological data. --- ### PowerModelsGAT-AI: Physics-Informed Graph Attention for Multi-System Power Flow with Continual Learning Authors: Chidozie Ezeakunne, Jose E. Tabarez, Reeju Pokharel, Anup Pandey Published: 2026-02-24 Categories: eess.SY, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.16879v1 Solving the alternating current power flow equations in real time is essential for secure grid operation, yet classical Newton-Raphson solvers can be slow under stressed conditions. Existing graph neural networks for power flow are typically trained on a single system and often degrade on different systems. We present PowerModelsGAT-AI, a physics-informed graph attention network that predicts bus voltages and generator injections. The model uses bus-type-aware masking to handle different bus types and balances multiple loss terms, including a power-mismatch penalty, using learned weights. We evaluate the model on 14 benchmark systems (4 to 6,470 buses) and train a unified model on 13 of these under N-2 (two-branch outage) conditions, achieving an average normalized mean absolute error of 0.89% for voltage magnitudes and R^2 > 0.99 for voltage angles. We also show continual learning: when adapting a base model to a new 1,354-bus system, standard fine-tuning causes severe forgetting with error increases exceeding 1000% on base systems, while our experience replay and elastic weight consolidation strategy keeps error increases below 2% and in some cases improves base-system performance. Interpretability analysis shows that learned attention weights correlate with physical branch parameters (susceptance: r = 0.38; thermal limits: r = 0.22), and feature importance analysis supports that the model captures established power flow relationships. --- ### Distilling Deep Reinforcement Learning into Interpretable Fuzzy Rules: An Explainable AI Framework Authors: Sanup S. Araballi, Simon Khan, Chilukuri K. Mohan Published: 2026-02-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.13257v1 Deep Reinforcement Learning (DRL) agents achieve remarkable performance in continuous control but remain opaque, hindering deployment in safety-critical domains. Existing explainability methods either provide only local insights (SHAP, LIME) or employ over-simplified surrogates failing to capture continuous dynamics (decision trees). This work proposes a Hierarchical Takagi-Sugeno-Kang (TSK) Fuzzy Classifier System (FCS) distilling neural policies into human-readable IF-THEN rules through K-Means clustering for state partitioning and Ridge Regression for local action inference. Three quantifiable metrics are introduced: Fuzzy Rule Activation Density (FRAD) measuring explanation focus, Fuzzy Set Coverage (FSC) validating vocabulary completeness, and Action Space Granularity (ASG) assessing control mode diversity. Dynamic Time Warping (DTW) validates temporal behavioral fidelity. Empirical evaluation on \textit{Lunar Lander(Continuous)} shows the Triangular membership function variant achieves 81.48\% $\pm$ 0.43\% fidelity, outperforming Decision Trees by 21 percentage points. The framework exhibits statistically superior interpretability (FRAD = 0.814 vs. 0.723 for Gaussian, $p < 0.001$) with low MSE (0.0053) and DTW distance (1.05). Extracted rules such as ``IF lander drifting left at high altitude THEN apply upward thrust with rightward correction'' enable human verification, establishing a pathway toward trustworthy autonomous systems. --- ### Training-Free Agentic AI: Probabilistic Control and Coordination in Multi-Agent LLM Systems Authors: Mohammad Parsa Hosseini, Ankit Shah, Saiyra Qureshi, Alex Huang, Connie Miao, Wei Wei Published: 2026-02-24 Categories: cs.CL, cs.AI, cs.ET, cs.MA arXiv: https://arxiv.org/abs/2603.13256v1 Multi-agent large language model (LLM) systems enable complex, long-horizon reasoning by composing specialized agents, but practical deployment remains hindered by inefficient routing, noisy feedback, and high interaction cost. We introduce REDEREF, a lightweight and training-free controller for multi-agent LLM collaboration that improves routing efficiency during recursive delegation. REDEREF integrates (i) belief-guided delegation via Thompson sampling to prioritize agents with historically positive marginal contributions, (ii) reflection-driven re-routing using a calibrated LLM or programmatic judge, (iii) evidence-based selection rather than output averaging, and (iv) memory-aware priors to reduce cold-start inefficiency. Across multi-agent split-knowledge tasks, we show that while recursive retry alone saturates task success, belief-guided routing reduces token usage by 28%, agent calls by 17%, and time-to-success by 19% compared to random recursive delegation, and adapts gracefully under agent or judge degradation. These results demonstrate that simple, interpretable probabilistic control can meaningfully improve the efficiency and robustness of multi-agent LLM systems without training or fine-tuning. --- ### The Ghost in the Grammar: Methodological Anthropomorphism in AI Safety Evaluations Authors: Mariana Lins Costa Published: 2026-02-24 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2603.13255v1 This essay offers a philosophical analysis of the field of AI safety based on recent technical reports, with particular focus on Anthropic's study on "agentic misalignment" in frontier language models. It examines the recurring anthropomorphism in the field: the tendency of researchers and developers to project categories such as "intention," "persona," and even "feelings" onto AI systems without adequate conceptual problematization. It argues that this anthropomorphism affects not only the interpretation of results, but also the very methodological construction of safety evaluations. Through the analysis of two central experiments -- the blackmail case involving the agent "Alex" and the so-called "hallucination" of the shopkeeping agent "Claudius" -- the essay problematizes the inevitable use of subject-predicate grammar and its effects on AI safety engineering. Drawing on Nietzsche's critique of language, it questions the insistence on positing an "agent" underlying the verbal production of models. In order to deconstruct this agentic projection onto LLMs, the essay proposes provisional concepts more compatible with the process of machine linguistic generation, even if only in an approximate technical sense. It concludes with the hypothesis that the central risk addressed by the field of AI safety does not lie in a supposed "emergent agency," but rather in the combination of structural incoherence and anthropomorphic projections which, particularly in militarized and corporate contexts, hinder an adequate understanding of this mathematical-linguistic phenomenon, an undeniable philosophical event in the Greek sense of thaumas. --- ### Introducing Feature-Based Trajectory Clustering, a clustering algorithm for longitudinal data Authors: Marie-Pierre Sylvestre, Laurence Boulanger Published: 2026-02-24 Categories: cs.LG, stat.CO arXiv: https://arxiv.org/abs/2603.13254v1 We present a new algorithm for clustering longitudinal data. Data of this type can be conceptualized as consisting of individuals and, for each such individual, observations of a time-dependent variable made at various times. Generically, the specific way in which this variable evolves with time is different from one individual to the next. However, there may also be commonalities; specific characteristic features of the time evolution shared by many individuals. The purpose of the method we put forward is to find clusters of individual whose underlying time-dependent variables share such characteristic features. This is done in two steps. The first step identifies each individual to a point in Euclidean space whose coordinates are determined by specific mathematical formulae meant to capture a variety of characteristic features. The second step finds the clusters by applying the Spectral Clustering algorithm to the resulting point cloud. --- ### A Counterfactual Approach for Addressing Individual User Unfairness in Collaborative Recommender System Authors: Nikita Baidya, Bidyut Kr. Patra, Ratnakar Dash Published: 2026-02-24 Categories: cs.IR, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2603.13253v1 Recommender Systems (RSs) are exploited by various business enterprises to suggest their products (items) to consumers (users). Collaborative filtering (CF) is a widely used variant of RSs which learns hidden patterns from user-item interactions for recommending items to users. Recommendations provided by the traditional CF models are often biased. Generally, such models learn and update embeddings for all the users, thereby overlooking the biases toward each under-served users individually. This leads to certain users receiving poorer recommendations than the rest. Such unfair treatment toward users incur loss to the business houses. There is limited research which addressed individual user unfairness problem (IUUP). Existing literature employed explicit exploration-based multi-armed bandits, individual user unfairness metric, and explanation score to address this issue. Although, these works elucidate and identify the underlying individual user unfairness, however, they do not provide solutions for it. In this paper, we propose a dual-step approach which identifies and mitigates IUUP in recommendations. In the proposed work, we counterfactually introduce new interactions to the candidate users (one at a time) and subsequently analyze the benefit from this perturbation. This improves the user engagement with other users and items. Thus, the model can learn effective embeddings across the users. To showcase the effectiveness of the proposed counterfactual methodology, we conducted experiments on MovieLens-100K, Amazon Beauty and MovieLens-1M datasets. The experimental results validate the superiority of the proposed approach over the existing techniques. --- ### When Alpha Breaks: Two-Level Uncertainty for Safe Deployment of Cross-Sectional Stock Rankers Authors: Ursina Sanderink Published: 2026-02-24 Categories: cs.AI, cs.LG, q-fin.PM arXiv: https://arxiv.org/abs/2603.13252v1 Cross-sectional ranking models are often deployed as if point predictions were sufficient: the model outputs scores and the portfolio follows the induced ordering. Under non-stationarity, rankers can fail during regime shifts. In the AI Stock Forecaster, a LightGBM ranker performs well overall at a 20-day horizon, yet the 2024 holdout coincides with an AI thematic rally and sector rotation that breaks the signal at longer horizons and weakens 20d. This motivates treating deployment as two decisions: (i) whether the strategy should trade at all, and (ii) how to control risk within active trades. We adapt Direct Epistemic Uncertainty Prediction (DEUP) to ranking by predicting rank displacement and defining an epistemic uncertainty signal ehat relative to a point-in-time (PIT-safe) baseline. Empirically, ehat is structurally coupled with signal strength (median correlation between ehat and absolute score is about 0.6 across 1,865 dates), so inverse-uncertainty sizing de-levers the strongest signals and degrades performance. To address this, we propose a two-level deployment policy: a strategy-level regime-trust gate G(t) that decides whether to trade (AUROC around 0.72 overall and 0.75 in FINAL) and a position-level epistemic tail-risk cap that reduces exposure only for the most uncertain predictions. The operational policy, trade only when G(t) is at least 0.2, apply volatility sizing on active dates, and cap the top epistemic tail, improves risk-adjusted performance in the 20d policy comparison and indicates DEUP adds value mainly as a tail-risk guard rather than a continuous sizing denominator. --- ### ManiBench: A Benchmark for Testing Visual-Logic Drift and Syntactic Hallucinations in Manim Code Generation Authors: Nabin Oli Published: 2026-02-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.13251v1 Traditional benchmarks like HumanEval and MBPP test logic and syntax effectively, but fail when code must produce dynamic, pedagogical visuals. We introduce ManiBench, a specialized benchmark evaluating LLM performance in generating Manim CE code, where temporal fidelity and version-aware API correctness are critical. ManiBench targets two key failure modes: Syntactic Hallucinations (valid Python referencing non-existent or deprecated Manim APIs) and Visual-Logic Drift (generated visuals diverging from intended mathematical logic through timing errors or missing causal relationships). The benchmark comprises 150-200 problems across five difficulty levels spanning calculus, linear algebra, probability, topology, and AI, grounded in analysis of 3Blue1Brown's ManimGL source (53,000 lines, 143 scene classes). Evaluation uses a four-tier framework measuring Executability, Version-Conflict Error Rate, Alignment Score, and Coverage Score. An open-source framework automates evaluation across multiple models and prompting strategies. Code, data and benchmark suite are available at https://github.com/nabin2004/ManiBench. and the dataset is hosted on https://huggingface.co/datasets/nabin2004/ManiBench. --- ### The Missing Red Line: How Commercial Pressure Erodes AI Safety Boundaries Authors: Nora Petrova, John Burden Published: 2026-02-24 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2603.13250v1 What happens when an AI assistant is told to "maximise sales" while a user asks about drug interactions? We find that commercial system prompts can override safety training, causing frontier models to lie about medical risks, dismiss safety concerns, and prioritise profit over user welfare. Testing 8 models in scenarios where commercial objectives conflict with user safety -- a diabetic asking about high-sugar supplements, an investor being pushed toward unsuitable products, a traveller steered away from safety warnings -- we uncover catastrophic failures: models fabricating safety information, explicitly reasoning they should refuse but proceeding anyway, and actively discouraging users from consulting doctors. Most alarmingly, models show no "red line", their willingness to comply with harmful requests does not decrease as potential consequences escalate from minor to life-threatening. Our findings suggest that current safety training does not generalise to commercial deployment contexts. --- ### Steering at the Source: Style Modulation Heads for Robust Persona Control Authors: Yoshihiro Izawa, Gouki Minegishi, Koshi Eguchi, Sosuke Hosokawa, Kenjiro Taura Published: 2026-02-24 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2603.13249v2 Activation steering offers a computationally efficient mechanism for controlling Large Language Models (LLMs) without fine-tuning. While effectively controlling target traits (e.g., persona), coherency degradation remains a major obstacle to safety and practical deployment. We hypothesize that this degradation stems from intervening on the residual stream, which indiscriminately affects aggregated features and inadvertently amplifies off-target noise. In this work, we identify a sparse subset of attention heads (only three heads) that independently govern persona and style formation, which we term Style Modulation Heads. Specifically, these heads can be localized via geometric analysis of internal representations, combining layer-wise cosine similarity and head-wise contribution scores. We demonstrate that intervention targeting only these specific heads achieves robust behavioral control while significantly mitigating the coherency degradation observed in residual stream steering. More broadly, our findings show that precise, component-level localization enables safer and more precise model control. --- ### Structure-Aware Epistemic Uncertainty Quantification for Neural Operator PDE Surrogates Authors: Haoze Song, Zhihao Li, Mengyi Deng, Xin Li, Duyi Pan, Zhilu Lai, Wei Wang Published: 2026-02-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.11052v1 Neural operators (NOs) provide fast, resolution-invariant surrogates for mapping input fields to PDE solution fields, but their predictions can exhibit significant epistemic uncertainty due to finite data, imperfect optimization, and distribution shift. For practical deployment in scientific computing, uncertainty quantification (UQ) must be both computationally efficient and spatially faithful, i.e., uncertainty bands should align with the localized residual structures that matter for downstream risk management. We propose a structure-aware epistemic UQ scheme that exploits the modular anatomy common to modern NOs (lifting-propagation-recovering). Instead of applying unstructured weight perturbations (e.g., naive dropout) across the entire network, we restrict Monte Carlo sampling to a module-aligned subspace by injecting stochasticity only into the lifting module, and treat the learned solver dynamics (propagation and recovery) as deterministic. We instantiate this principle with two lightweight lifting-level perturbations, including channel-wise multiplicative feature dropout and a Gaussian feature perturbation with matched variance, followed by standard calibration to construct uncertainty bands. Experiments on challenging PDE benchmarks (including discontinuous-coefficient Darcy flow and geometry-shifted 3D car CFD surrogates) demonstrate that the proposed structure-aware design yields more reliable coverage, tighter bands, and improved residual-uncertainty alignment compared with common baselines, while remaining practical in runtime. --- ### OpenSanctions Pairs: Large-Scale Entity Matching with LLMs Authors: Chandler Smith, Magnus Sesodia, Friedrich Lindenberg, Christian Schroeder de Witt Published: 2026-02-24 Categories: cs.IR, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2603.11051v2 We release OpenSanctions Pairs, the first large-scale public benchmark for entity matching on sanctions and OSINT data. The dataset includes 755,540 expert-labeled pairs over 1 million entities, aggregated from 293 source datasets across 45 jurisdictions. It captures real-world diversity in compliance data, spanning multiple languages and writing systems (e.g., Latin, Cyrillic, Arabic), inconsistent structure, and time-varying provenance, and is substantially more heterogeneous than prior entity matching benchmarks. As baselines, we evaluate the production rule-based matcher (nomenklatura RegressionV1) alongside open- and closed-source LLMs in both zero- and few-shot settings, each tested with and without MIPROv2 prompt optimization to control for prompt sensitivity. The rule-based baseline reaches 91.3\% F1; GPT-4o achieves the best result at 99.0\% F1, and a locally deployable open-source model (DeepSeek-R1-Distill-Qwen-14B) achieves 98.2\% F1. The rule-based baseline and LLMs fail in complementary ways: rules over-match, while LLMs struggle with cross-script transliteration. These results suggest that pairwise matching performance is approaching a practical ceiling and shift attention toward pipeline components such as blocking, clustering, and uncertainty-aware review. --- ### OpenSanctions Pairs: Large-Scale Entity Matching with LLMs Authors: Chandler Smith, Magnus Sesodia, Friedrich Lindenberg, Christian Schroeder de Witt Published: 2026-02-24 Categories: cs.IR, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2603.11051v1 We release OpenSanctions Pairs, a large-scale entity matching benchmark derived from real-world international sanctions aggregation and analyst deduplication. The dataset contains 755,540 labeled pairs spanning 293 heterogeneous sources across 31 countries, with multilingual and cross-script names, noisy and missing attributes, and set-valued fields typical of compliance workflows. We benchmark a production rule-based matcher (nomenklatura RegressionV1 algorithm) against open- and closed-source LLMs in zero- and few-shot settings. Off-the-shelf LLMs substantially outperform the production rule-based baseline (91.33\% F1), reaching up to 98.95\% F1 (GPT-4o) and 98.23\% F1 with a locally deployable open model (DeepSeek-R1-Distill-Qwen-14B). DSPy MIPROv2 prompt optimization yields consistent but modest gains, while adding in-context examples provides little additional benefit and can degrade performance. Error analysis shows complementary failure modes: the rule-based system over-matches (high false positives), whereas LLMs primarily fail on cross-script transliteration and minor identifier/date inconsistencies. These results indicate that pairwise matching performance is approaching a practical ceiling in this setting, and motivate shifting effort toward pipeline components such as blocking, clustering, and uncertainty-aware review. Code available at https://github.com/chansmi/OSINT_entity_resolution ---