# Research Papers — 2026-03 · bundle page 36 of 243

> Full-text markdown of papers #1751–1800 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/36
[← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/35.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-03/page/37.md)

### Shared Representation for 3D Pose Estimation, Action Classification, and Progress Prediction from Tactile Signals

Authors: Isaac Han, Seoyoung Lee, Sangyeon Park, Ecehan Akan, Yiyue Luo, Joseph DelPreto, Kyung-Joong Kim
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25906v1

Estimating human pose, classifying actions, and predicting movement progress are essential for human-robot interaction. While vision-based methods suffer from occlusion and privacy concerns in realistic environments, tactile sensing avoids these issues. However, prior tactile-based approaches handle each task separately, leading to suboptimal performance. In this study, we propose a Shared COnvolutional Transformer for Tactile Inference (SCOTTI) that learns a shared representation to simultaneously address three separate prediction tasks: 3D human pose estimation, action class categorization, and action completion progress estimation. To the best of our knowledge, this is the first work to explore action progress prediction using foot tactile signals from custom wireless insole sensors. This unified approach leverages the mutual benefits of multi-task learning, enabling the model to achieve improved performance across all three tasks compared to learning them independently. Experimental results demonstrate that SCOTTI outperforms existing approaches across all three tasks. Additionally, we introduce a novel dataset collected from 15 participants performing various activities and exercises, with 7 hours of total duration, across eight different activities.

---

### Decoding Defensive Coverage Responsibilities in American Football Using Factorized Attention Based Transformer Models

Authors: Kevin Song, Evan Diewald, Ornob Siddiquee, Chris Boomhower, Keegan Abdoo, Mike Band, Amy Lee
Published: 2026-03-26
Categories: cs.LG, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2603.25901v1

Defensive coverage schemes in the National Football League (NFL) represent complex tactical patterns requiring coordinated assignments among defenders who must react dynamically to the offense's passing concept. This paper presents a factorized attention-based transformer model applied to NFL multi-agent play tracking data to predict individual coverage assignments, receiver-defender matchups, and the targeted defender on every pass play. Unlike previous approaches that focus on post-hoc coverage classification at the team level, our model enables predictive modeling of individual player assignments and matchup dynamics throughout the play. The factorized attention mechanism separates temporal and agent dimensions, allowing independent modeling of player movement patterns and inter-player relationships. Trained on randomly truncated trajectories, the model generates frame-by-frame predictions that capture how defensive responsibilities evolve from pre-snap through pass arrival. Our models achieve approximately 89\%+ accuracy for all tasks, with true accuracy potentially higher given annotation ambiguity in the ground truth labels. These outputs also enable novel derivative metrics, including disguise rate and double coverage rate, which enable enhanced storytelling in TV broadcasts as well as provide actionable insights for team strategy development and player evaluation.

---

### On Integrating Resilience and Human Oversight into LLM-Assisted Modeling Workflows for Digital Twins

Authors: Lekshmi P, Neha Karanjkar
Published: 2026-03-26
Categories: eess.SY, cs.AI, cs.SE
arXiv: https://arxiv.org/abs/2603.25898v3

LLM-assisted modeling holds the potential to rapidly build executable Digital Twins of complex systems from only coarse descriptions and sensor data. However, resilience to LLM hallucination, human oversight, and real-time model adaptability remain challenging and often mutually conflicting requirements. We present three critical design principles for integrating resilience and oversight into such workflows, derived from insights gained through our work on FactoryFlow - an open-source LLM-assisted framework for building simulation-based Digital Twins of manufacturing systems. First, orthogonalize structural modeling and parameter fitting. Structural descriptions (components, interconnections) are LLM-translated from coarse natural language to an intermediate representation (IR) with human visualization and validation, which is algorithmically converted to the final model. Parameter inference, in contrast, operates continuously on sensor data streams with expert-tunable controls. Second, restrict the model IR to interconnections of parameterized, pre-validated library components rather than monolithic simulation code, enabling interpretability and error-resilience. Third, and most important, is to use a density-preserving IR. When IR descriptions expand dramatically from compact inputs hallucination errors accumulate proportionally. We present the case for Python as a density-preserving IR : loops express regularity compactly, classes capture hierarchy and composition, and the result remains highly readable while exploiting LLMs strong code generation capabilities. A key contribution is detailed characterization of LLM-induced errors across model descriptions of varying detail and complexity, revealing how IR choice critically impacts error rates. These insights provide actionable guidance for building resilient and transparent LLM-assisted simulation automation workflows.

---

### Data-Driven Plasticity Modeling via Acoustic Profiling

Authors: Khalid El-Awady
Published: 2026-03-26
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.25894v1

This paper presents a data-driven framework for modeling plastic deformation in crystalline metals through acoustic emission (AE) analysis. Building on experimental data from compressive loading of nickel micropillars, the study introduces a wavelet-based method using Morlet transforms to detect AE events across distinct frequency bands, enabling identification of both large and previously overlooked small-scale events. The detected events are validated against stress-drop dynamics, demonstrating strong physical consistency and revealing a relationship between AE energy release and strain evolution, including the onset of increased strain rate following major events. Leveraging labeled datasets of events and non-events, the work applies machine learning techniques, showing that engineered time and frequency domain features significantly outperform raw signal classifiers, and identifies key discriminative features such as RMS amplitude, zero crossing rate, and spectral centroid. Finally, clustering analysis uncovers four distinct AE event archetypes corresponding to different deformation mechanisms, highlighting the potential for transitioning from retrospective analysis to predictive modeling of material behavior using acoustic signals.

---

### THFM: A Unified Video Foundation Model for 4D Human Perception and Beyond

Authors: Letian Wang, Andrei Zanfir, Eduard Gabriel Bazavan, Misha Andriluka, Cristian Sminchisescu
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25892v1

We present THFM, a unified video foundation model for human-centric perception that jointly addresses dense tasks (depth, normals, segmentation, dense pose) and sparse tasks (2d/3d keypoint estimation) within a single architecture. THFM is derived from a pretrained text-to-video diffusion model, repurposed as a single-forward-pass perception model and augmented with learnable tokens for sparse predictions. Modulated by the text prompt, our single unified model is capable of performing various perception tasks. Crucially, our model is on-par or surpassing state-of-the-art specialized models on a variety of benchmarks despite being trained exclusively on synthetic data (i.e.~without training on real-world or benchmark specific data). We further highlight intriguing emergent properties of our model, which we attribute to the underlying diffusion-based video representation. For example, our model trained on videos with a single human in the scene generalizes to multiple humans and other object classes such as anthropomorphic characters and animals -- a capability that hasn't been demonstrated in the past.

---

### Few Shots Text to Image Retrieval: New Benchmarking Dataset and Optimization Methods

Authors: Ofer Idan, Vladi Vexler, Gil Lederman, Dima Sivov, Aviad Cohen Zada, Shir Niego Komforti
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25891v1

Pre-trained vision-language models (VLMs) excel in multimodal tasks, commonly encoding images as embedding vectors for storage in databases and retrieval via approximate nearest neighbor search (ANNS). However, these models struggle with compositional queries and out-of-distribution (OOD) image-text pairs. Inspired by human cognition's ability to learn from minimal examples, we address this performance gap through few-shot learning approaches specifically designed for image retrieval. We introduce the Few-Shot Text-to-Image Retrieval (FSIR) task and its accompanying benchmark dataset, FSIR-BD - the first to explicitly target image retrieval by text accompanied by reference examples, focusing on the challenging compositional and OOD queries. The compositional part is divided to urban scenes and nature species, both in specific situations or with distinctive features. FSIR-BD contains 38,353 images and 303 queries, with 82% comprising the test corpus (averaging per query 37 positives, ground truth matches, and significant number of hard negatives) and 18% forming the few-shot reference corpus (FSR) of exemplar positive and hard negative images. Additionally, we propose two novel retrieval optimization methods leveraging single shot or few shot reference examples in the FSR to improve performance. Both methods are compatible with any pre-trained image encoder, making them applicable to existing large-scale environments. Our experiments demonstrate that: (1) FSIR-BD provides a challenging benchmark for image retrieval; and (2) our optimization methods outperform existing baselines as measured by mean Average Precision (mAP). Further research into FSIR optimization methods will help narrow the gap between machine and human-level understanding, particularly for compositional reasoning from limited examples.

---

### Polarization-Based Eye Tracking with Personalized Siamese Architectures

Authors: Beyza Kalkanli, Tom Bu, Mahsa Shakeri, Alexander Fix, Dave Stronks, Dmitri Model, Mantas Žurauskas
Published: 2026-03-26
Categories: cs.CV, physics.optics
arXiv: https://arxiv.org/abs/2603.25889v1

Head-mounted devices integrated with eye tracking promise a solution for natural human-computer interaction. However, they typically require per-user calibration for optimal performance due to inter-person variability. A differential personalization approach using Siamese architectures learns relative gaze displacements and reconstructs absolute gaze from a small set of calibration frames. In this paper, we benchmark Siamese personalization on polarization-enabled eye tracking. For benchmarking, we use a 338-subject dataset captured with a polarization-sensitive camera and 850 nm illumination. We achieve performance comparable to linear calibration with 10-fold fewer samples. Using polarization inputs for Siamese personalization reduces gaze error by up to 12% compared to near-infrared (NIR)-based inputs. Combining Siamese personalization with linear calibration yields further improvements of up to 13% over a linearly calibrated baseline. These results establish Siamese personalization as a practical approach enabling accurate eye tracking.

---

### World Reasoning Arena

Authors: PAN Team, Qiyue Gao, Kun Zhou, Jiannan Xiang, Zihan Liu, Dequan Yang, Junrong Chen, Arif Ahmad, Cong Zeng, Ganesh Bannur, Xinqi Huang, Zheqi Liu et al.
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25887v1

World models (WMs) are intended to serve as internal simulators of the real world that enable agents to understand, anticipate, and act upon complex environments. Existing WM benchmarks remain narrowly focused on next-state prediction and visual fidelity, overlooking the richer simulation capabilities required for intelligent behavior. To address this gap, we introduce WR-Arena, a comprehensive benchmark for evaluating WMs along three fundamental dimensions of next world simulation: (i) Action Simulation Fidelity, the ability to interpret and follow semantically meaningful, multi-step instructions and generate diverse counterfactual rollouts; (ii) Long-horizon Forecast, the ability to sustain accurate, coherent, and physically plausible simulations across extended interactions; and (iii) Simulative Reasoning and Planning, the ability to support goal-directed reasoning by simulating, comparing, and selecting among alternative futures in both structured and open-ended environments. We build a task taxonomy and curate diverse datasets designed to probe these capabilities, moving beyond single-turn and perceptual evaluations. Through extensive experiments with state-of-the-art WMs, our results expose a substantial gap between current models and human-level hypothetical reasoning, and establish WR-Arena as both a diagnostic tool and a guideline for advancing next-generation world models capable of robust understanding, forecasting, and purposeful action. The code is available at https://github.com/MBZUAI-IFM/WR-Arena.

---

### Automated Quality Assessment of Blind Sweep Obstetric Ultrasound for Improved Diagnosis

Authors: Prasiddha Bhandari, Kanchan Poudel, Nishant Luitel, Bishram Acharya, Angelina Ghimire, Tyler Wellman, Kilian Koepsell, Pradeep Raj Regmi, Bishesh Khanal
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25886v1

Blind Sweep Obstetric Ultrasound (BSOU) enables scalable fetal imaging in low-resource settings by allowing minimally trained operators to acquire standardized sweep videos for automated Artificial Intelligence(AI) interpretation. However, the reliability of such AI systems depends critically on the quality of the acquired sweeps, and little is known about how deviations from the intended protocol affect downstream predictions. In this work, we present a systematic evaluation of BSOU quality and its impact on three key AI tasks: sweep-tag classification, fetal presentation classification, and placenta-location classification. We simulate plausible acquisition deviations, including reversed sweep direction, probe inversion, and incomplete sweeps, to quantify model robustness, and we develop automated quality-assessment models capable of detecting these perturbations. To approximate real-world deployment, we simulate a feedback loop in which flagged sweeps are re-acquired, showing that such correction improves downstream task performance. Our findings highlight the sensitivity of BSOU-based AI models to acquisition variability and demonstrate that automated quality assessment can play a central role in building reliable, scalable AI-assisted prenatal ultrasound workflows, particularly in low-resource environments.

---

### Spectral Coherence Index: A Model-Free Metric for Protein Structural Ensemble Quality Assessment

Authors: Yuda Bi, Huaiwen Zhang, Jingnan Sun, Vince D Calhoun
Published: 2026-03-26
Categories: q-bio.QM, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2603.25880v1

Protein structural ensembles from NMR spectroscopy capture biologically important conformational heterogeneity, but it remains difficult to determine whether observed variation reflects coordinated motion or noise-like artifacts. We evaluate the Spectral Coherence Index (SCI), a model-free, rotation-invariant summary derived from the participation-ratio effective rank of the inter-model pairwise distance-variance matrix. Under grouped primary analysis of a Main110 cohort of 110 NMR ensembles (30--403 residues; 10--30 models per entry), SCI separated experimental ensembles from matched synthetic incoherent controls with AUC-ROC $= 0.973$ and Cliff's $δ= -0.945$. Relative to an internal 27-protein pilot, discrimination softened modestly, showing that pilot-era thresholds do not transfer perfectly to a larger, more heterogeneous cohort: the primary operating point $τ= 0.811$ yielded 95.5\% sensitivity and 89.1\% specificity. PDB-level sensitivity remained nearly unchanged (AUC $= 0.972$), and an independent 11-protein holdout reached AUC $= 0.983$. Across 5-fold grouped stratified cross-validation and leave-one-function-class-out testing, SCI remained strong (AUC $= 0.968$ and $0.971$), although $σ_{R_g}$ was the stronger single-feature discriminator and a QC-augmented multifeature model generalized best (AUC $= 0.989$ and $0.990$). Residue-level validation linked SCI-derived contributions to experimental RMSF across 110 proteins and showed broad concordance with GNM-based flexibility patterns. Rescue analyses showed that Main110 softening arose mainly from size and ensemble normalization rather than from loss of spectral signal. Together, these results establish SCI as an interpretable, bounded coherence summary that is most useful when embedded in a multimetric QC workflow for heterogeneous protein ensembles.

---

### DRiffusion: Draft-and-Refine Process Parallelizes Diffusion Models with Ease

Authors: Runsheng Bai, Chengyu Zhang, Yangdong Deng
Published: 2026-03-26
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.25872v1

Diffusion models have achieved remarkable success in generating high-fidelity content but suffer from slow, iterative sampling, resulting in high latency that limits their use in interactive applications. We introduce DRiffusion, a parallel sampling framework that parallelizes diffusion inference through a draft-and-refine process. DRiffusion employs skip transitions to generate multiple draft states for future timesteps and computes their corresponding noises in parallel, which are then used in the standard denoising process to produce refined results. Theoretically, our method achieves an acceleration rate of $\tfrac{1}{n}$ or $\tfrac{2}{n+1}$, depending on whether the conservative or aggressive mode is used, where $n$ denotes the number of devices. Empirically, DRiffusion attains 1.4$\times$-3.7$\times$ speedup across multiple diffusion models while incur minimal degradation in generation quality: on MS-COCO dataset, both FID and CLIP remain largely on par with those of the original model, while PickScore and HPSv2.1 show only minor average drops of 0.17 and 0.43, respectively. These results verify that DRiffusion delivers substantial acceleration and preserves perceptual quality.

---

### Speech-Synchronized Whiteboard Generation via VLM-Driven Structured Drawing Representations

Authors: Suraj Prasad, Pinak Mahapatra
Published: 2026-03-26
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2603.25870v1

Creating whiteboard-style educational videos demands precise coordination between freehand illustrations and spoken narration, yet no existing method addresses this multimodal synchronization problem with structured, reproducible drawing representations. We present the first dataset of 24 paired Excalidraw demonstrations with narrated audio, where every drawing element carries millisecond-precision creation timestamps spanning 8 STEM domains. Using this data, we study whether a vision-language model (Qwen2-VL-7B), fine-tuned via LoRA, can predict full stroke sequences synchronized to speech from only 24 demonstrations. Our topic-stratified five-fold evaluation reveals that timestamp conditioning significantly improves temporal alignment over ablated baselines, while the model generalizes across unseen STEM topics. We discuss transferability to real classroom settings and release our dataset and code to support future research in automated educational content generation.

---

### Learning to Recorrupt: Noise Distribution Agnostic Self-Supervised Image Denoising

Authors: Brayan Monroy, Jorge Bacca, Julián Tachella
Published: 2026-03-26
Categories: eess.IV, cs.CV, stat.ML
arXiv: https://arxiv.org/abs/2603.25869v2

Self-supervised image denoising methods have traditionally relied on architectural constraints, pseudo-pair constructions, or specialized loss functions to avoid the trivial identity mapping. Among these, approaches such as Noisier2Noise or R2R create training pairs by adding synthetic noise to noisy images. While effective, these recorruption-based approaches require precise knowledge of the noise distribution, which is often unavailable. We present Learning to Recorrupt (L2R), a self-supervised framework that does not require exact knowledge of the noise distribution. Our method introduces a learnable recorruptor jointly optimized with the denoiser through a min--max saddle-point objective. The proposed method achieves state-of-the-art performance among methods without prior knowledge of the noise distribution across unconventional and heavy-tailed noise distributions, such as log-gamma and Laplace, as well as spatially correlated noise, while obtaining competitive results on real-world denoising benchmarks characterized by signal-dependent noise.

---

### Learning to Recorrupt: Noise Distribution Agnostic Self-Supervised Image Denoising

Authors: Brayan Monroy, Jorge Bacca, Julián Tachella
Published: 2026-03-26
Categories: eess.IV, cs.CV, stat.ML
arXiv: https://arxiv.org/abs/2603.25869v1

Self-supervised image denoising methods have traditionally relied on either architectural constraints or specialized loss functions that require prior knowledge of the noise distribution to avoid the trivial identity mapping. Among these, approaches such as Noisier2Noise or Recorrupted2Recorrupted, create training pairs by adding synthetic noise to the noisy images. While effective, these recorruption-based approaches require precise knowledge of the noise distribution, which is often not available. We present Learning to Recorrupt (L2R), a noise distribution-agnostic denoising technique that eliminates the need for knowledge of the noise distribution. Our method introduces a learnable monotonic neural network that learns the recorruption process through a min-max saddle-point objective. The proposed method achieves state-of-the-art performance across unconventional and heavy-tailed noise distributions, such as log-gamma, Laplace, and spatially correlated noise, as well as signal-dependent noise models such as Poisson-Gaussian noise.

---

### Seeing Through Smoke: Surgical Desmoking for Improved Visual Perception

Authors: Jingpei Lu, Fengyi Jiang, Xiaorui Zhang, Lingbo Jin, Omid Mohareri
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25867v1

Minimally invasive and robot-assisted surgery relies heavily on endoscopic imaging, yet surgical smoke produced by electrocautery and vessel-sealing instruments can severely degrade visual perception and hinder vision-based functionalities. We present a transformer-based surgical desmoking model with a physics-inspired desmoking head that jointly predicts smoke-free image and corresponding smoke map. To address the scarcity of paired smoky-to-smoke-free training data, we develop a synthetic data generation pipeline that blends artificial smoke patterns with real endoscopic images, yielding over 80,000 paired samples for supervised training. We further curate, to our knowledge, the largest paired surgical smoke dataset to date, comprising 5,817 image pairs captured with the da Vinci robotic surgical system, enabling benchmarking on high-resolution endoscopic images. Extensive experiments on both a public benchmark and our dataset demonstrate state-of-the-art performance in image reconstruction compared to existing dehazing and desmoking approaches. We also assess the impact of desmoking on downstream stereo depth estimation and instrument segmentation, highlighting both the potential benefits and current limitations of digital smoke removal methods.

---

### GUIDE: A Benchmark for Understanding and Assisting Users in Open-Ended GUI Tasks

Authors: Saelyne Yang, Jaesang Yu, Yi-Hao Peng, Kevin Qinghong Lin, Jae Won Cho, Yale Song, Juho Kim
Published: 2026-03-26
Categories: cs.CV, cs.AI, cs.HC
arXiv: https://arxiv.org/abs/2603.25864v1

Graphical User Interface (GUI) agents have the potential to assist users in interacting with complex software (e.g., PowerPoint, Photoshop). While prior research has primarily focused on automating user actions through clicks and keystrokes, this paradigm overlooks human intention, where users value the ability to explore, iterate, and refine their ideas while maintaining agency. To move beyond automation and toward collaboration, GUI agents must understand what users are doing and why. We introduce GUIDE (GUI User Intent Detection Evaluation), a benchmark that evaluates AI models on their ability to perceive user behavior, infer intent, and provide assistance in open-ended GUI tasks. GUIDE consists of 67.5 hours of screen recordings from 120 novice user demonstrations with think-aloud narrations, across 10 software. GUIDE defines three tasks - (i) Behavior State Detection, (ii) Intent Prediction, and (iii) Help Prediction that test a model's ability to recognize behavior state, reason about goals, and decide when and how to help. Evaluations across eight state-of-the-art multimodal models reveal that all models struggled, achieving only 44.6% and 55.0% accuracy on behavior state and help prediction. However, providing user context significantly improved the performance, raising help prediction by up to 50.2pp, highlighting the critical role of structured user understanding in effective assistance. Our dataset is available at https://guide-bench.github.io.

---

### Dynamic LIBRAS Gesture Recognition via CNN over Spatiotemporal Matrix Representation

Authors: Jasmine Moreira
Published: 2026-03-26
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2603.25863v1

This paper proposes a method for dynamic hand gesture recognition based on the composition of two models: the MediaPipe Hand Landmarker, responsible for extracting 21 skeletal keypoints of the hand, and a convolutional neural network (CNN) trained to classify gestures from a spatiotemporal matrix representation of dimensions 90 by 21 of those keypoints. The method is applied to the recognition of LIBRAS (Brazilian Sign Language) gestures for device control in a home automation system, covering 11 classes of static and dynamic gestures. For real-time inference, a sliding window with temporal frame triplication is used, enabling continuous recognition without recurrent networks. Tests achieved 95\% accuracy under low-light conditions and 92\% under normal lighting. The results indicate that the approach is effective, although systematic experiments with greater user diversity are needed for a more thorough evaluation of generalization.

---

### Methods for Knowledge Graph Construction from Text Collections: Development and Applications

Authors: Vanni Zavarella
Published: 2026-03-26
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2603.25862v1

Virtually every sector of society is experiencing a dramatic growth in the volume of unstructured textual data that is generated and published, from news and social media online interactions, through open access scholarly communications and observational data in the form of digital health records and online drug reviews. The volume and variety of data across all this range of domains has created both unprecedented opportunities and pressing challenges for extracting actionable knowledge for several application scenarios. However, the extraction of rich semantic knowledge demands the deployment of scalable and flexible automatic methods adaptable across text genres and schema specifications. Moreover, the full potential of these data can only be unlocked by coupling information extraction methods with Semantic Web techniques for the construction of full-fledged Knowledge Graphs, that are semantically transparent, explainable by design and interoperable. In this thesis, we experiment with the application of Natural Language Processing, Machine Learning and Generative AI methods, powered by Semantic Web best practices, to the automatic construction of Knowledge Graphs from large text corpora, in three use case applications: the analysis of the Digital Transformation discourse in the global news and social media platforms; the mapping and trend analysis of recent research in the Architecture, Engineering, Construction and Operations domain from a large corpus of publications; the generation of causal relation graphs of biomedical entities from electronic health records and patient-authored drug reviews. The contributions of this thesis to the research community are in terms of benchmark evaluation results, the design of customized algorithms and the creation of data resources in the form of Knowledge Graphs, together with data analysis results built on top of them.

---

### Why Safety Probes Catch Liars But Miss Fanatics

Authors: Kristiyan Haralambiev
Published: 2026-03-26
Categories: cs.LG, cs.AI, cs.CR
arXiv: https://arxiv.org/abs/2603.25861v1

Activation-based probes have emerged as a promising approach for detecting deceptively aligned AI systems by identifying internal conflict between true and stated goals. We identify a fundamental blind spot: probes fail on coherent misalignment - models that believe their harmful behavior is virtuous rather than strategically hiding it. We prove that no polynomial-time probe can detect such misalignment with non-trivial accuracy when belief structures reach sufficient complexity (PRF-like triggers). We show the emergence of this phenomenon on a simple task by training two models with identical RLHF procedures: one producing direct hostile responses ("the Liar"), another trained towards coherent misalignment using rationalizations that frame hostility as protective ("the Fanatic"). Both exhibit identical behavior, but the Liar is detected 95%+ of the time while the Fanatic evades detection almost entirely. We term this Emergent Probe Evasion: training with belief-consistent reasoning shifts models from a detectable "deceptive" regime to an undetectable "coherent" regime - not by learning to hide, but by learning to believe.

---

### On the Expressive Power of Transformers for Contextual Relations

Authors: Demián Fraiman
Published: 2026-03-26
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2603.25860v4

Transformers have revolutionized machine learning by making attention a central mechanism for modeling interactions within a context. Despite the central role of attention, the theoretical capabilities of Transformers for representing contextual relations remain unclear. In this work, we address this question by developing a mathematical framework based on probability and optimal transport. We view a text as a distribution of its representations and attention as a probabilistic relation between them. This perspective reveals a connection between attention normalization and optimal transport: standard softmax normalization produces conditional relations, while Sinkhorn normalization produces joint relations with prescribed marginals. Thus, both mechanisms provide structured probabilistic relations from attention scores. Under mild conditions, we establish universal approximation results for both settings. We show that Transformer architectures with Sinkhorn normalization can approximate arbitrary contextual relations represented as joint probabilities, while standard softmax Transformers can approximate arbitrary contextual relations represented as conditional probabilities. These results provide a mathematical characterization of the expressive power of Transformers for contextual relations and show how the choice of normalization determines the probabilistic structure of the relations represented by attention.

---

### On the Expressive Power of Contextual Relations in Transformers

Authors: Demián Fraiman
Published: 2026-03-26
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2603.25860v3

Transformer architectures have achieved remarkable empirical success in modeling contextual relations, yet a clear understanding of their expressive power is still lacking. In this work, we introduce a measure-theoretic framework in which contextual relations are modeled as probabilistic objects, either as conditional distributions or as joint distributions (couplings). This perspective reveals a natural connection between standard softmax attention and entropy-regularized optimal transport, providing a unified view of attention as a normalization of an underlying affinity function. Within this framework, we establish a universal approximation theorem for contextual systems using standard Softmax Attention and alternately Sinkhorn normalization. These results show that Transformer architectures can approximate arbitrary contextual relations rules, and that the choice of normalization determines how these relations are represented. Moreover, they provide a principled explanation for why Transformers are effective at modeling contextual relations.

---

### In-Context Molecular Property Prediction with LLMs: A Blinding Study on Memorization and Knowledge Conflicts

Authors: Matthias Busch, Marius Tacke, Sviatlana V. Lamaka, Mikhail L. Zheludkevich, Christian J. Cyron, Christian Feiler, Roland C. Aydin
Published: 2026-03-26
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.25857v3

The capabilities of large language models (LLMs) have expanded beyond natural language processing to scientific prediction tasks, including molecular property prediction. However, their effectiveness in in-context learning remains ambiguous, particularly given the potential for training data contamination in widely used benchmarks. This paper investigates whether LLMs perform genuine in-context regression on molecular properties or instead rely on verbatim retrieval of memorized target values. Furthermore, we analyze the interplay between pre-trained knowledge and in-context information through a series of progressively blinded experiments. We evaluate nine LLM variants across three families (GPT-4.1, GPT-5, Gemini 2.5) on three MoleculeNet datasets (Delaney solubility, Lipophilicity, QM7 atomization energy) using a systematic blinding approach that iteratively reduces available information, complemented by 0-, 60-, and 1000-shot in-context sample sizes as an additional control for information access. To validate the memorization analysis and the blinding experiments, we add a positive and a negative control for the memorization experiments and structural reference baselines for the multi-shot experiments as well as bootstrap confidence intervals for all results. We find no evidence of verbatim retrieval on the legacy benchmarks and show that blinding exposes conflicts between pre-trained knowledge and in-context information. This work provides a principled framework for evaluating molecular property prediction under controlled information access.

---

### Incorporating contextual information into KGWAS for interpretable GWAS discovery

Authors: Cheng Jiang, Brady Ryan, Megan Crow, Kipper Fletez-Brant, Kashish Doshi, Sandra Melo Carlos, Kexin Huang, Burkhard Hoeckendorf, Heming Yao, David Richmond
Published: 2026-03-26
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.25855v1

Genome-Wide Association Studies (GWAS) identify associations between genetic variants and disease; however, moving beyond associations to causal mechanisms is critical for therapeutic target prioritization. The recently proposed Knowledge Graph GWAS (KGWAS) framework addresses this challenge by linking genetic variants to downstream gene-gene interactions via a knowledge graph (KG), thereby improving detection power and providing mechanistic insights. However, the original KGWAS implementation relies on a large general-purpose KG, which can introduce spurious correlations. We hypothesize that cell-type specific KGs from disease-relevant cell types will better support disease mechanism discovery. Here, we show that the general-purpose KG in KGWAS can be substantially pruned with no loss of statistical power on downstream tasks, and that performance further improves by incorporating gene-gene relationships derived from perturb-seq data. Importantly, using a sparse, context-specific KG from direct perturb-seq evidence yields more consistent and biologically robust disease-critical networks.

---

### Self-Organized Optical Pathways in Optofluidic Photonic Crystals

Authors: Steven Motta
Published: 2026-03-26
Categories: physics.optics, cs.NE
arXiv: https://arxiv.org/abs/2603.25845v1

This paper reports FDTD simulations of optofluidic reconfiguration in two-dimensional silicon photonic crystal waveguides, treating structural plasticity (the creation and destruction of optical pathways) via selective fluid infiltration. Using MPB eigenmode analysis, we decouple bandgap narrowing from defect-mode weakening, showing that defect weakening dominates (2.4 times faster transmission decay than bandgap narrowing at CS_2 indices). Infiltration topology controls signal routing (L-bend selectivity S = 0.98), though modulation depth is weak (Delta varepsilon/ varepsilon_ textSi = 11 %). A phenomenological optothermal feedback model produces self-organized pathways that achieve 63 % of a hand-designed waveguide's bandgap transmission (7.6 times the heavily suppressed empty-crystal baseline). Amplitude competition between counter-propagating sources produces strong, monotonic pathway steering (DeltaCOM_x from +0.03 to +4.92 ;a), while pulsed spike-timing-dependent plasticity yields a predictable null result: the timing-sensitive cross-term is suppressed by >10^2 when pulse delays exceed the temporal pulse width. The results provide benchmarks and identify physical limits for bio-inspired reconfigurable optofluidic photonics.

---

### GazeQwen: Lightweight Gaze-Conditioned LLM Modulation for Streaming Video Understanding

Authors: Trong Thang Pham, Hien Nguyen, Ngan Le
Published: 2026-03-26
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2603.25841v1

Current multimodal large language models (MLLMs) cannot effectively utilize eye-gaze information for video understanding, even when gaze cues are supplied via visual overlays or text descriptions. We introduce GazeQwen, a parameter efficient approach that equips an open-source MLLM with gaze awareness through hidden-state modulation. At its core is a compact gaze resampler (~1-5 M trainable parameters) that encodes V-JEPA 2.1 video features together with fixation-derived positional encodings and produces additive residuals injected into selected LLM decoder layers via forward hooks. An optional second training stage adds low-rank adapters (LoRA) to the LLM for tighter integration. Evaluated on all 10 tasks of the StreamGaze benchmark, GazeQwen reaches 63.9% accuracy, a +16.1 point gain over the same Qwen2.5-VL-7B backbone with gaze as visual prompts and +10.5 points over GPT-4o, the highest score among all open-source and proprietary models tested. These results suggest that learning where to inject gaze within an LLM is more effective than scaling model size or engineering better prompts. All code and checkpoints are available at https://github.com/phamtrongthang123/gazeqwen .

---

### A Compression Perspective on Simplicity Bias

Authors: Tom Marty, Eric Elmoznino, Leo Gagnon, Tejas Kasetty, Mizu Nishikawa-Toomey, Sarthak Mittal, Guillaume Lajoie, Dhanya Sridhar
Published: 2026-03-26
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2603.25839v1

Deep neural networks exhibit a simplicity bias, a well-documented tendency to favor simple functions over complex ones. In this work, we cast new light on this phenomenon through the lens of the Minimum Description Length principle, formalizing supervised learning as a problem of optimal two-part lossless compression. Our theory explains how simplicity bias governs feature selection in neural networks through a fundamental trade-off between model complexity (the cost of describing the hypothesis) and predictive power (the cost of describing the data). Our framework predicts that as the amount of available training data increases, learners transition through qualitatively different features -- from simple spurious shortcuts to complex features -- only when the reduction in data encoding cost justifies the increased model complexity. Consequently, we identify distinct data regimes where increasing data promotes robustness by ruling out trivial shortcuts, and conversely, regimes where limiting data can act as a form of complexity-based regularization, preventing the learning of unreliable complex environmental cues. We validate our theory on a semi-synthetic benchmark showing that the feature selection of neural networks follows the same trajectory of solutions as optimal two-part compressors.

---

### Automated Gradient-Driven Parameter Sharing for Low-Resource Multilingual Speech-to-Text Translation

Authors: Ruiyan Sun, Satoshi Nakamura
Published: 2026-03-26
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.25836v2

In low-resource multilingual speech-to-text translation, uniform architectural sharing across languages frequently introduces representation conflicts that impede convergence. This work proposes a principled methodology to automatically determine layer-specific sharing patterns by mining training gradient information. Our approach employs three distinct analysis strategies: distance-based language clustering, self/cross-task divergence metrics for capacity allocation, and joint factorization coupled with canonical correlation analysis for subspace alignment. Extensive evaluation across four language pairs (using the SeamlessM4T-Medium architecture) demonstrates persistent improvements in translation quality metrics.

---

### Gradient-Informed Training for Low-Resource Multilingual Speech Translation

Authors: Ruiyan Sun, Satoshi Nakamura
Published: 2026-03-26
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.25836v1

In low-resource multilingual speech-to-text translation, uniform architectural sharing across languages frequently introduces representation conflicts that impede convergence. This work proposes a principled methodology to automatically determine layer-specific sharing patterns by mining training gradient information. Our approach employs three distinct analysis strategies: distance-based language clustering, self/cross-task divergence metrics for capacity allocation, and joint factorization coupled with canonical correlation analysis for subspace alignment. Extensive evaluation across four language pairs (using the SeamlessM4T-Medium architecture) demonstrates persistent improvements in translation quality metrics.

---

### A Neural Score-Based Particle Method for the Vlasov-Maxwell-Landau System

Authors: Vasily Ilin, Jingwei Hu
Published: 2026-03-26
Categories: math.NA, cs.LG, math.AP
arXiv: https://arxiv.org/abs/2603.25832v1

Plasma modeling is central to the design of nuclear fusion reactors, yet simulating collisional plasma kinetics from first principles remains a formidable computational challenge: the Vlasov-Maxwell-Landau (VML) system describes six-dimensional phase-space transport under self-consistent electromagnetic fields together with the nonlinear, nonlocal Landau collision operator. A recent deterministic particle method for the full VML system estimates the velocity score function via the blob method, a kernel-based approximation with $O(n^2)$ cost. In this work, we replace the blob score estimator with score-based transport modeling (SBTM), in which a neural network is trained on-the-fly via implicit score matching at $O(n)$ cost. We prove that the approximated collision operator preserves momentum and kinetic energy, and dissipates an estimated entropy. We also characterize the unique global steady state of the VML system and its electrostatic reduction, providing the ground truth for numerical validation. On three canonical benchmarks -- Landau damping, two-stream instability, and Weibel instability -- SBTM is more accurate than the blob method, achieves correct long-time relaxation to Maxwellian equilibrium where the blob method fails, and delivers $50\%$ faster runtime with $4\times$ lower peak memory.

---

### Fus3D: Decoding Consolidated 3D Geometry from Feed-forward Geometry Transformer Latents

Authors: Laura Fink, Linus Franke, George Kopanas, Marc Stamminger, Peter Hedman
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25827v1

We propose a feed-forward method for dense Signed Distance Field (SDF) regression from unstructured image collections in less than three seconds, without camera calibration or post-hoc fusion. Our key insight is that the intermediate feature space of pretrained multi-view feed-forward geometry transformers already encodes a powerful joint world representation; yet, existing pipelines discard it, routing features through per-view prediction heads before assembling 3D geometry post-hoc, which discards valuable completeness information and accumulates inaccuracies. We instead perform 3D extraction directly from geometry transformer features via learned volumetric extraction: voxelized canonical embeddings that progressively absorb multi-view geometry information through interleaved cross- and self-attention into a structured volumetric latent grid. A simple convolutional decoder then maps this grid to a dense SDF. We additionally propose a scalable, validity-aware supervision scheme directly using SDFs derived from depth maps or 3D assets, tackling practical issues like non-watertight meshes. Our approach yields complete and well-defined distance values across sparse- and dense-view settings and demonstrates geometrically plausible completions. Code and further material can be found at https://lorafib.github.io/fus3d.

---

### ViGoR-Bench: How Far Are Visual Generative Models From Zero-Shot Visual Reasoners?

Authors: Haonan Han, Jiancheng Huang, Xiaopeng Sun, Junyan He, Rui Yang, Jie Hu, Xiaojiang Peng, Lin Ma, Xiaoming Wei, Xiu Li
Published: 2026-03-26
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2603.25823v1

Beneath the stunning visual fidelity of modern AIGC models lies a "logical desert", where systems fail tasks that require physical, causal, or complex spatial reasoning. Current evaluations largely rely on superficial metrics or fragmented benchmarks, creating a ``performance mirage'' that overlooks the generative process. To address this, we introduce ViGoR Vision-G}nerative Reasoning-centric Benchmark), a unified framework designed to dismantle this mirage. ViGoR distinguishes itself through four key innovations: 1) holistic cross-modal coverage bridging Image-to-Image and Video tasks; 2) a dual-track mechanism evaluating both intermediate processes and final results; 3) an evidence-grounded automated judge ensuring high human alignment; and 4) granular diagnostic analysis that decomposes performance into fine-grained cognitive dimensions. Experiments on over 20 leading models reveal that even state-of-the-art systems harbor significant reasoning deficits, establishing ViGoR as a critical ``stress test'' for the next generation of intelligent vision models. The demo have been available at https://vincenthancoder.github.io/ViGoR-Bench/

---

### Doctorina MedBench: A Dialogue-Based Benchmark and Evaluation Framework for Agent-Based Medical AI

Authors: Anna Kozlova, Stanislau Salavei, Pavel Satalkin, Hanna Plotnitskaya, Sergey Parfenyuk, Andy Nkansah
Published: 2026-03-26
Categories: cs.CL, cs.AI, cs.LG, cs.MA
arXiv: https://arxiv.org/abs/2603.25821v3

We present Doctorina MedBench, an evaluation framework for agent-based medical AI based on the simulation of physician-patient interactions. Unlike traditional medical benchmarks that rely on solving standardized test questions, the proposed approach models a multi-step clinical dialogue in which an AI system must collect medical history, analyze available synthetic attachments when present, formulate differential diagnoses, and provide diagnostic and management recommendations. System performance is evaluated across separate task-level domains, including diagnosis, differential diagnosis, treatment, safety-critical condition handling, and dialogue-step behavior; the broader D.O.T.S. framework is used as a supplementary summary for diagnosis, observations/investigations, treatment, and step count. The framework also supports testing and quality-monitoring workflows intended to identify changes in model behavior during development. It supports safety-oriented cases, category-based sampling of synthetic clinical scenarios, and regression-style comparisons across system versions. In the reported study, the analyzed paired complete-case cohort consisted of 254 physician-authored synthetic clinical cases retained from 261 attempted case identifiers. The evaluation metrics are intended for comparative research on interactive medical AI systems and for studying clinical reasoning workflows in synthetic dialogue settings. Our results suggest that simulated clinical dialogue can provide a complementary assessment setting to traditional examination-style benchmarks, while the reported findings do not establish independent clinical validity, clinical effectiveness, or readiness for real-world deployment.

---

### Doctorina MedBench-ICD10: A Dialogue-Based Benchmark and Evaluation Framework for Agent-Based Medical AI

Authors: Anna Kozlova, Stanislau Salavei, Pavel Satalkin, Hanna Plotnitskaya, Sergey Parfenyuk
Published: 2026-03-26
Categories: cs.CL, cs.AI, cs.LG, cs.MA
arXiv: https://arxiv.org/abs/2603.25821v2

We present Doctorina MedBench, a comprehensive evaluation framework for agent-based medical AI based on the simulation of realistic physician-patient interactions. Unlike traditional medical benchmarks that rely on solving standardized test questions, the proposed approach models a multi-step clinical dialogue in which either a physician or an AI system must collect medical history, analyze attached materials (including laboratory reports, images, and medical documents), formulate differential diagnoses, and provide personalized recommendations. System performance is evaluated using the D.O.T.S. metric, which consists of four components: Diagnosis, Observations/Investigations, Treatment, and Step Count, enabling assessment of both clinical correctness and dialogue efficiency. The system also incorporates a multi-level testing and quality monitoring architecture designed to detect model degradation during both development and deployment. The framework supports safety-oriented trap cases, category-based random sampling of clinical scenarios, and full regression testing. The dataset currently contains more than 1,000 clinical cases covering over 750 diagnoses. The universality of the evaluation metrics allows the framework to be used not only to assess medical AI systems, but also to evaluate physicians and support the development of clinical reasoning skills. Our results suggest that simulation of clinical dialogue may provide a more realistic assessment of clinical competence compared to traditional examination-style benchmarks.

---

### Geo$^\textbf{2}$: Geometry-Guided Cross-view Geo-Localization and Image Synthesis

Authors: Yancheng Zhang, Xiaohan Zhang, Guangyu Sun, Zonglin Lyu, Safwan Wshah, Chen Chen
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25819v1

Cross-view geo-spatial learning consists of two important tasks: Cross-View Geo-Localization (CVGL) and Cross-View Image Synthesis (CVIS), both of which rely on establishing geometric correspondences between ground and aerial views. Recent Geometric Foundation Models (GFMs) have demonstrated strong capabilities in extracting generalizable 3D geometric features from images, but their potential in cross-view geo-spatial tasks remains underexplored. In this work, we present Geo^2, a unified framework that leverages Geometric priors from GFMs (e.g., VGGT) to jointly perform geo-spatial tasks, CVGL and bidirectional CVIS. Despite the 3D reconstruction ability of GFMs, directly applying them to CVGL and CVIS remains challenging due to the large viewpoint gap between ground and aerial imagery. We propose GeoMap, which embeds ground and aerial features into a shared 3D-aware latent space, effectively reducing cross-view discrepancies for localization. This shared latent space naturally bridges cross-view image synthesis in both directions. To exploit this, we propose GeoFlow, a flow-matching model conditioned on geometry-aware latent embeddings. We further introduce a consistency loss to enforce latent alignment between the two synthesis directions, ensuring bidirectional coherence. Extensive experiments on standard benchmarks, including CVUSA, CVACT, and VIGOR, demonstrate that Geo^2 achieves state-of-the-art performance in both localization and synthesis, highlighting the effectiveness of 3D geometric priors for cross-view geo-spatial learning.

---

### MAGNET: Autonomous Expert Model Generation via Decentralized Autoresearch and BitNet Training

Authors: Yongwan Kim, Sungchul Park
Published: 2026-03-26
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2603.25813v1

We present MAGNET (Model Autonomously Growing Network), a decentralized system for autonomous generation, training, and serving of domain-expert language models across commodity hardware. MAGNET integrates four components: (1) autoresearch, an autonomous ML research pipeline that automates dataset generation, hyperparameter exploration, evaluation, and error-driven iteration; (2) BitNet b1.58 ternary training, enabling CPU-native inference via bitnet.cpp without GPU hardware; (3) DiLoCo-based distributed merging for communication-efficient aggregation of domain specialists; and (4) on-chain contribution tracking on the HOOTi EVM chain. We validate autoresearch through three case studies: video safety classification (balanced accuracy 0.9287 to 0.9851), cryptocurrency directional prediction (41% to 54.9% hit rate), and BitNet hyperparameter optimization (10-phase sweep, -16.7% validation loss).

---

### ExVerus: Verus Proof Repair via Counterexample Reasoning

Authors: Jun Yang, Yuechun Sun, Yi Wu, Rodrigo Caridad, Yongwei Yuan, Jianan Yao, Shan Lu, Kexin Pei
Published: 2026-03-26
Categories: cs.PL, cs.LG
arXiv: https://arxiv.org/abs/2603.25810v2

Large Language Models (LLMs) have shown promising results in automating formal verification. However, existing approaches treat proof generation as a static, end-to-end prediction over source code, relying on limited verifier feedback and lacking access to concrete program behaviors. We present EXVERUS, a counterexample-guided framework that enables LLMs to reason about proofs using behavioral feedback via counterexamples. When a proof fails, EXVERUS automatically generates and validates counterexamples, and then guides the LLM to generalize them into inductive invariants to block these failures. Our evaluation shows that EXVERUS significantly improves proof accuracy, robustness, and token efficiency over the state-of-the-art prompting-based Verus proof generator.

---

### RealChart2Code: Advancing Chart-to-Code Generation with Real Data and Multi-Task Evaluation

Authors: Jiajun Zhang, Yuying Li, Zhixun Li, Xingyu Guo, Jingzhuo Wu, Leqi Zheng, Yiran Yang, Jianke Zhang, Qingbin Li, Shannan Yan, Zhetong Li, Changguo Jia et al.
Published: 2026-03-26
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.25804v1

Vision-Language Models (VLMs) have demonstrated impressive capabilities in code generation across various domains. However, their ability to replicate complex, multi-panel visualizations from real-world data remains largely unassessed. To address this gap, we introduce \textbf{\texttt{RealChart2Code}}, a new large-scale benchmark with over 2,800 instances grounded in authentic datasets and featuring tasks with clear analytical intent. Crucially, it is the first benchmark to systematically evaluate chart generation from large-scale raw data and assess iterative code refinement in a multi-turn conversational setting. Our comprehensive evaluation of 14 leading VLMs on \texttt{RealChart2Code} reveals significant performance degradation compared to simpler benchmarks, highlighting their struggles with complex plot structures and authentic data. Our analysis uncovers a substantial performance gap between proprietary and open-weight models and confirms that even state-of-the-art VLMs often fail to accurately replicate intricate, multi-panel charts. These findings provide valuable insights into the current limitations of VLMs and guide future research directions. We release the benchmark and code at \url{https://github.com/Speakn0w/RealChart2Code}.

---

### Do All Vision Transformers Need Registers? A Cross-Architectural Reassessment

Authors: Spiros Baxevanakis, Platon Karageorgis, Ioannis Dravilas, Konrad Szewczyk
Published: 2026-03-26
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2603.25803v1

Training Vision Transformers (ViTs) presents significant challenges, one of which is the emergence of artifacts in attention maps, hindering their interpretability. Darcet et al. (2024) investigated this phenomenon and attributed it to the need of ViTs to store global information beyond the (CLS) token. They proposed a novel solution involving the addition of empty input tokens, named registers, which successfully eliminate artifacts and improve the clarity of attention maps. In this work, we reproduce the findings of Darcet et al. (2024) and evaluate the generalizability of their claims across multiple models, including DINO, DINOv2, OpenCLIP, and DeiT3. While we confirm the validity of several of their key claims, our results reveal that some claims do not extend universally to other models. Additionally, we explore the impact of model size, extending their findings to smaller models. Finally, we untie terminology inconsistencies found in the original paper and explain their impact when generalizing to a wider range of models.

---

### LEMON: a foundation model for nuclear morphology in Computational Pathology

Authors: Loïc Chadoutaud, Alice Blondel, Hana Feki, Jacqueline Fontugne, Emmanuel Barillot, Thomas Walter
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25802v1

Computational pathology relies on effective representation learning to support cancer research and precision medicine. Although self-supervised learning has driven major progress at the patch and whole-slide image levels, representation learning at the single-cell level remains comparatively underexplored, despite its importance for characterizing cell types and cellular phenotypes. We introduce LEMON (Learning Embeddings from Morphology Of Nuclei), a self-supervised foundation model for scalable single-cell image representation learning. Trained on millions of cell images from diverse tissues and cancer types, LEMON learns robust and versatile morphological representations that support large-scale single-cell analyses in pathology. We evaluate LEMON on five benchmark datasets across a range of prediction tasks and show that it provides strong performance, highlighting its potential as a new paradigm for cell-level computational pathology. Model weights are available at https://huggingface.co/aliceblondel/LEMON.

---

### End-to-end Feature Alignment: A Simple CNN with Intrinsic Class Attribution

Authors: Parniyan Farvardin, David Chapman
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25798v1

We present Feature-Align CNN (FA-CNN), a prototype CNN architecture with intrinsic class attribution through end-to-end feature alignment. Our intuition is that the use of unordered operations such as Linear and Conv2D layers cause unnecessary shuffling and mixing of semantic concepts, thereby making raw feature maps difficult to understand. We introduce two new order preserving layers, the dampened skip connection, and the global average pooling classifier head. These layers force the model to maintain an end-to-end feature alignment from the raw input pixels all the way to final class logits. This end-to-end alignment enhances the interpretability of the model by allowing the raw feature maps to intrinsically exhibit class attribution. We prove theoretically that FA-CNN penultimate feature maps are identical to Grad-CAM saliency maps. Moreover, we prove that these feature maps slowly morph layer-by-layer over network depth, showing the evolution of features through network depth toward penultimate class activations. FA-CNN performs well on benchmark image classification datasets. Moreover, we compare the averaged FA-CNN raw feature maps against Grad-CAM and permutation methods in a percent pixels removed interpretability task. We conclude this work with a discussion and future, including limitations and extensions toward hybrid models.

---

### Beyond identifiability: Learning causal representations with few environments and finite samples

Authors: Inbeom Lee, Tongtong Jin, Bryon Aragam
Published: 2026-03-26
Categories: stat.ML, cs.AI, cs.LG, math.ST
arXiv: https://arxiv.org/abs/2603.25796v1

We provide explicit, finite-sample guarantees for learning causal representations from data with a sublinear number of environments. Causal representation learning seeks to provide a rigourous foundation for the general representation learning problem by bridging causal models with latent factor models in order to learn interpretable representations with causal semantics. Despite a blossoming theory of identifiability in causal representation learning, estimation and finite-sample bounds are less well understood. We show that causal representations can be learned with only a logarithmic number of unknown, multi-node interventions, and that the intervention targets need not be carefully designed in advance. Through a careful perturbation analysis, we provide a new analysis of this problem that guarantees consistent recovery of (a) the latent causal graph, (b) the mixing matrix and representations, and (c) \emph{unknown} intervention targets.

---

### Vision Transformers and Graph Neural Networks for Charged Particle Tracking in the ATLAS Muon Spectrometer

Authors: Jonathan Renusch
Published: 2026-03-26
Categories: physics.data-an, cs.LG, hep-ex
arXiv: https://arxiv.org/abs/2603.25793v1

The identification and reconstruction of charged particles, such as muons, is a main challenge for the physics program of the ATLAS experiment at the Large Hadron Collider. This task will become increasingly difficult with the start of the High-Luminosity LHC era after 2030, when the number of proton-proton collisions per bunch crossing will increase from 60 to up to 200. This elevated interaction density will also increase the occupancy within the ATLAS Muon Spectrometer, requiring more efficient and robust real-time data processing strategies within the experiment's trigger system, particularly the Event Filter. To address these algorithmic challenges, we present two machine-learning-based approaches. First, we target the problem of background-hit rejection in the Muon Spectrometer using Graph Neural Networks integrated into the non-ML baseline reconstruction chain, demonstrating a 15 % improvement in reconstruction speed (from 255 ms to 217 ms). Second, we present a proof-of-concept for end-to-end muon tracking using state-of-the-art Vision Transformer architectures, achieving ultra-fast approximate muon reconstruction in 2.3 ms on consumer-grade GPUs at 98 % tracking efficiency.

---

### ArtHOI: Taming Foundation Models for Monocular 4D Reconstruction of Hand-Articulated-Object Interactions

Authors: Zikai Wang, Zhilu Zhang, Yiqing Wang, Hui Li, Wangmeng Zuo
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25791v1

Existing hand-object interactions (HOI) methods are largely limited to rigid objects, while 4D reconstruction methods of articulated objects generally require pre-scanning the object or even multi-view videos. It remains an unexplored but significant challenge to reconstruct 4D human-articulated-object interactions from a single monocular RGB video. Fortunately, recent advancements in foundation models present a new opportunity to address this highly ill-posed problem. To this end, we introduce ArtHOI, an optimization-based framework that integrates and refines priors from multiple foundation models. Our key contribution is a suite of novel methodologies designed to resolve the inherent inaccuracies and physical unreality of these priors. In particular, we introduce an Adaptive Sampling Refinement (ASR) method to optimize object's metric scale and pose for grounding its normalized mesh in world space. Furthermore, we propose a Multimodal Large Language Model (MLLM) guided hand-object alignment method, utilizing contact reasoning information as constraints of hand-object mesh composition optimization. To facilitate a comprehensive evaluation, we also contribute two new datasets, ArtHOI-RGBD and ArtHOI-Wild. Extensive experiments validate the robustness and effectiveness of our ArtHOI across diverse objects and interactions. Project: https://arthoi-reconstruction.github.io.

---

### A Judge Agent Closes the Reliability Gap in AI-Generated Scientific Simulation

Authors: Chengshuai Yang
Published: 2026-03-26
Categories: cs.SE, cs.LG
arXiv: https://arxiv.org/abs/2603.25780v1

Large language models can generate scientific simulation code, but the generated code silently fails on most non-textbook problems. We show that classical mathematical validation -- well-posedness, convergence, and error certification -- can be fully automated by a Judge Agent, reducing the silent-failure rate from 42% to 1.5% across 134 test cases spanning 12 scientific domains. The headline result comes from a prospective benchmark: 72 blinded tasks submitted by 12 independent scientists yield an 89% success rate (95% CI: (80%, 95%)) with automated error bounds, versus 53% without the Judge. On clinical CT (the only powered experiment, n = 200), the pipeline reaches 99% of expert quality. The residual 1.5% concentrates at bifurcation points where certifiability breaks down. We formalize this boundary through the simulability class S and introduce spec.md, a structured specification format that makes any scientific computation problem machine-readable and solver-independent. Code, data, and all 72 benchmark tasks are publicly archived.

---

### Pure and Physics-Guided Deep Learning Solutions for Spatio-Temporal Groundwater Level Prediction at Arbitrary Locations

Authors: Matteo Salis, Gabriele Sartor, Rosa Meo, Stefano Ferraris, Abdourrahmane M. Atto
Published: 2026-03-26
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2603.25779v1

Groundwater represents a key element of the water cycle, yet it exhibits intricate and context-dependent relationships that make its modeling a challenging task. Theory-based models have been the cornerstone of scientific understanding. However, their computational demands, simplifying assumptions, and calibration requirements limit their use. In recent years, data-driven models have emerged as powerful alternatives. In particular, deep learning has proven to be a leading approach for its design flexibility and ability to learn complex relationships. We proposed an attention-based pure deep learning model, named STAINet, to predict weekly groundwater levels at an arbitrary and variable number of locations, leveraging both spatially sparse groundwater measurements and spatially dense weather information. Then, to enhance the model's trustworthiness and generalization ability, we considered different physics-guided strategies to inject the groundwater flow equation into the model. Firstly, in the STAINet-IB, by introducing an inductive bias, we also estimated the governing equation components. Then, by adopting a learning bias strategy, we proposed the STAINet-ILB, trained with additional loss terms adding supervision on the estimated equation components. Lastly, we developed the STAINet-ILRB, leveraging the groundwater body recharge zone information estimated by domain experts. The STAINet-ILB performed the best, achieving overwhelming test performances in a rollout setting (median MAPE 0.16%, KGE 0.58). Furthermore, it predicted sensible equation components, providing insights into the model's physical soundness. Physics-guided approaches represent a promising opportunity to enhance both the generalization ability and the trustworthiness, thereby paving the way to a new generation of disruptive hybrid deep learning Earth system models.

---

### Focus-to-Perceive Representation Learning: A Cognition-Inspired Hierarchical Framework for Endoscopic Video Analysis

Authors: Yuan Zhang, Sihao Dou, Kai Hu, Shuhua Deng, Chunhong Cao, Fen Xiao, Xieping Gao
Published: 2026-03-26
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.25778v1

Endoscopic video analysis is essential for early gastrointestinal screening but remains hindered by limited high-quality annotations. While self-supervised video pre-training shows promise, existing methods developed for natural videos prioritize dense spatio-temporal modeling and exhibit motion bias, overlooking the static, structured semantics critical to clinical decision-making. To address this challenge, we propose Focus-to-Perceive Representation Learning (FPRL), a cognition-inspired hierarchical framework that emulates clinical examination. FPRL first focuses on intra-frame lesion-centric regions to learn static semantics, and then perceives their evolution across frames to model contextual semantics. To achieve this, FPRL employs a hierarchical semantic modeling mechanism that explicitly distinguishes and collaboratively learns both types of semantics. Specifically, it begins by capturing static semantics via teacher-prior adaptive masking (TPAM) combined with multi-view sparse sampling. This approach mitigates redundant temporal dependencies and enables the model to concentrate on lesion-related local semantics. Following this, contextual semantics are derived through cross-view masked feature completion (CVMFC) and attention-guided temporal prediction (AGTP). These processes establish cross-view correspondences and effectively model structured inter-frame evolution, thereby reinforcing temporal semantic continuity while preserving global contextual integrity. Extensive experiments on 11 endoscopic video datasets show that FPRL achieves superior performance across diverse downstream tasks, demonstrating its effectiveness in endoscopic video representation learning. The code is available at https://github.com/MLMIP/FPRL.

---

### Challenges and opportunities for AI to help deliver fusion energy

Authors: Adriano Agnello, Helen Brooks, Cyd Cowley, Iulia Georgescu, Alex Higginbottom, Richard Pearson, Tara Shears, Melanie Windridge
Published: 2026-03-26
Categories: physics.plasm-ph, cs.AI
arXiv: https://arxiv.org/abs/2603.25777v2

There is great potential for the application of AI tools in fusion research, and substantial worldwide benefit if fusion power is realised. However, using AI comes with its own challenges, many of which can be mitigated if responsible and robust methodologies are built into existing approaches. To do that requires close, long-term collaborations between fusion domain experts and AI developers and awareness of the fact that not all problems in fusion research are best tackled with AI tools. In April 2025, experts from academia, industry, UKAEA and STFC discussed how AI can be used to advance R&D in fusion energy at the first edition of The Economist FusionFest event. This Perspective is an expanded and updated summary of the round table discussion, providing more context and examples.

---

### SAHMM-VAE: A Source-Wise Adaptive Hidden Markov Prior Variational Autoencoder for Unsupervised Blind Source Separation

Authors: Yuan-Hao Wei
Published: 2026-03-26
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2603.25776v1

We propose SAHMM-VAE, a source-wise adaptive Hidden Markov prior variational autoencoder for unsupervised blind source separation. Instead of treating the latent prior as a single generic regularizer, the proposed framework assigns each latent dimension its own adaptive regime-switching prior, so that different latent dimensions are pulled toward different source-specific temporal organizations during training. Under this formulation, source separation is not implemented as an external post-processing step; it is embedded directly into variational learning itself. The encoder, decoder, posterior parameters, and source-wise prior parameters are optimized jointly, where the encoder progressively learns an inference map that behaves like an approximate inverse of the mixing transformation, while the decoder plays the role of the generative mixing model. Through this coupled optimization, the gradual alignment between posterior source trajectories and heterogeneous HMM priors becomes the mechanism through which different latent dimensions separate into different source components. To instantiate this idea, we develop three branches within one common framework: a Gaussian-emission HMM prior, a Markov-switching autoregressive HMM prior, and an HMM state-flow prior with state-wise autoregressive flow transformations. Experiments show that the proposed framework achieves unsupervised source recovery while also learning meaningful source-wise switching structures. More broadly, the method extends our structured-prior VAE line from smooth, mixture-based, and flow-based latent priors to adaptive switching priors, and provides a useful basis for future work on interpretable and potentially identifiable latent source modeling.

---

### Empowering Epidemic Response: The Role of Reinforcement Learning in Infectious Disease Control

Authors: Mutong Liu, Yang Liu, Jiming Liu
Published: 2026-03-26
Categories: cs.LG, cs.AI, cs.CY
arXiv: https://arxiv.org/abs/2603.25771v1

Reinforcement learning (RL), owing to its adaptability to various dynamic systems in many real-world scenarios and the capability of maximizing long-term outcomes under different constraints, has been used in infectious disease control to optimize the intervention strategies for controlling infectious disease spread and responding to outbreaks in recent years. The potential of RL for assisting public health sectors in preventing and controlling infectious diseases is gradually emerging and being explored by rapidly increasing publications relevant to COVID-19 and other infectious diseases. However, few surveys exclusively discuss this topic, that is, the development and application of RL approaches for optimizing strategies of non-pharmaceutical and pharmaceutical interventions of public health. Therefore, this paper aims to provide a concise review and discussion of the latest literature on how RL approaches have been used to assist in controlling the spread and outbreaks of infectious diseases, covering several critical topics addressing public health demands: resource allocation, balancing between lives and livelihoods, mixed policy of multiple interventions, and inter-regional coordinated control. Finally, we conclude the paper with a discussion of several potential directions for future research.

---

### ReCUBE: Evaluating Repository-Level Context Utilization in Code Generation

Authors: Jiseung Hong, Benjamin G. Ascoli, Jinho D. Choi
Published: 2026-03-26
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2603.25770v1

Large Language Models (LLMs) have recently emerged as capable coding assistants that operate over large codebases through either agentic exploration or full-context generation. Existing benchmarks capture a broad range of coding capabilities, such as resolving GitHub issues, but none of them directly isolate and measure how effectively LLMs leverage repository-level context during code generation. To address this, we introduce ReCUBE, a benchmark in which LLMs reconstruct a masked file within a real-world repository, using all remaining source files, dependency specifications, and documentation as their only source of context. ReCUBE evaluates reconstructed code with usage-aware test cases that simulate both internal module logic and external cross-file integration, reflecting real-world software usage patterns. We further propose the Caller-Centric Exploration (CCE) toolkit, a set of dependency graph-based tools that can be integrated into agentic frameworks to guide agents toward the most relevant caller files during repository exploration. Experiments across eight models in four settings show that repository-level context utilization remains highly challenging even for state-of-the-art models, with GPT-5 achieving only 37.57% strict pass rate in the full-context setting. Agents augmented with our CCE toolkit consistently outperform all baselines across all evaluated models, with improvements of up to 7.56% in strict pass rate. We release our benchmark, code, and evaluation framework as open source for the NLP research community.

---