research

Transformers Linearly Represent Highly Structured World Models

arXiv:2605.18847v1 Announce Type: cross Abstract: Do transformers, when trained on sequential reasoning traces, build internal models of the underlying task? And if so, does the structure of those int

Transforming Constraint Programs to Input for Local Search

DGX agent

arXiv:2605.19671v1 Announce Type: new Abstract: Applying local search algorithms to combinatorial optimization problems is not an easy feat. Typically, human intervention is required to compile the co

researcharxiv-cs-ai

20 May 2026

Trump has successfully censored most major US media. The US ranks #64 on the World Press Freedom index, which is below freedom. And it is ge…

Trump has successfully censored most major US media. The US ranks #64 on the World Press Freedom index, which is below freedom. And it is getting worse very soon. Sad to see that the US lacks checks a

researchyann-lecun--x

Trust It or Not: Evidential Uncertainty for Feed-Forward 3D Reconstruction with Trust3R

arXiv:2605.19539v1 Announce Type: new Abstract: Geometric foundation models hold promise for unconstrained dense geometry prediction from uncalibrated images. However, in current feed-forward designs,

Understanding Dynamics of Adam in Zero-Sum Games: An ODE Approach

arXiv:2605.19392v1 Announce Type: new Abstract: The remarkable success of the Adam in training neural networks has naturally led to the widespread use of its descent-ascent counterpart, Adam-DA, for s

UniRefiner: Teaching Pre-trained ViTs to Self-Dispose Dross via Contrastive Register

arXiv:2605.19622v1 Announce Type: new Abstract: Representation learning with Vision Transformers (ViTs) has advanced rapidly, yet the utility of large-scale models in spatially sensitive tasks is hind

Unsupervised Unfolded rPCA (U2-rPCA): Deep Interpretable Clutter Filtering for Ultrasound Microvascular Imaging

arXiv:2510.00660v2 Announce Type: replace Abstract: High-sensitivity clutter filtering is a fundamental step in ultrasound microvascular imaging. Singular value decomposition (SVD) and robust principa

Variance-Reduced Manifold Sampling via Polynomial-Maximization Density Estimation

arXiv:2605.19938v1 Announce Type: cross Abstract: Uniform sampling on implicitly defined manifolds is a core primitive in motion planning, constrained simulation, and probabilistic machine learning. M

Walsh-Hadamard Neural Operators for Solving PDEs with Discontinuous Coefficients

arXiv:2511.07347v2 Announce Type: replace-cross Abstract: Neural operators have emerged as powerful tools for learning solution operators of partial differential equations (PDEs). However, standard sp

Warm-starting active-set solvers using graph neural networks

arXiv:2511.13174v2 Announce Type: replace Abstract: Quadratic programming (QP) solvers are widely used in real-time control and optimization, but their computational cost often limits applicability in

WBCAtt+: Fine-Grained Pixel-Level Morphological Annotations for White Blood Cell Images

arXiv:2605.19692v1 Announce Type: new Abstract: The microscopic examination of white blood cells (WBCs) plays a fundamental role in pathology and is essential for diagnosing blood disorders such as le

What Are LLMs Doing to Scientific Communication? Measuring Changes in Writing Practices and Reading Experience

arXiv:2605.19936v1 Announce Type: new Abstract: Has the style of scientific communication changed due to the growing use of large language models in the writing process? We address this question in th

researcharxiv-cs-cl

What Machine Learning Can Do for Focusing Aerogel Detectors

arXiv:2312.02652v2 Announce Type: replace-cross Abstract: Particle identification at the Super Charm-Tau factory experiment will be provided by a Focusing Aerogel Ring Imaging CHerenkov detector (FARI

What Really Improves Mathematical Reasoning: Structured Reasoning Signals Beyond Pure Code

arXiv:2605.19762v1 Announce Type: new Abstract: Code has become a standard component of modern foundation language model (LM) training, yet its role beyond programming remains unclear. We revisit the

When Does Model Collapse Occur in Structured Interactive Learning?

arXiv:2605.20151v1 Announce Type: new Abstract: The proliferation of generative artificial intelligence has given rise to an interactive learning environment, where model parameters are continuously u

When the Majority Votes Wrong, the Intervention Timing for Test-Time Reinforcement Learning Hides in the Extinction Window

arXiv:2605.19444v1 Announce Type: cross Abstract: Test-time reinforcement learning (TTRL) reports substantial accuracy gains on mathematical reasoning benchmarks using majority vote as a pseudo-label

Where Does Authorship Signal Emerge in Encoder-Based Language Models?

arXiv:2605.19908v1 Announce Type: new Abstract: Authorship attribution models fine-tuned with the same pretrained encoder, data, and loss can differ four-fold in performance depending only on their sc

researcharxiv-cs-cl

White-Balance First, Adjust Later: Cross-Camera Color Constancy via Vision-Language Evaluation

arXiv:2605.19613v1 Announce Type: new Abstract: Color constancy aims to keep object colors consistent under varying illumination. Cross-camera generalization in color constancy remains challenging bec

Within one 24 hour period, Trump: - got out of a $100 million IRS fine - secured 'immunity' from all future tax investigations for his famil…

Within one 24 hour period, Trump: - got out of a 100 million IRS fine - secured 'immunity' from all future tax investigations for his family and friends - created a 1.8 billion slush fund for lawbreak

researchyann-lecun--x

X-Ray cardiac angiographic vessel segmentation based on pixel classification using machine learning and region growing

arXiv:2605.20073v1 Announce Type: new Abstract: This work proposes a pixel-classification approach for vessel segmentation in x-ray angiograms. The proposal uses textural features such as anisotropic

XFlowMap: Cross-Scale Generalization and Mapping of Massive Origin-Destination Data

arXiv:2605.18777v1 Announce Type: cross Abstract: Mapping large origin-destination (OD) datasets remains challenging because flow maps become cluttered, meaningful patterns occur at multiple spatial s

3D Densification for Multi-Map Monocular VSLAM in Endoscopy

arXiv:2503.14346v3 Announce Type: replace Abstract: Multi-map Sparse Monocular visual Simultaneous Localization and Mapping applied to monocular endoscopic sequences has proven efficient to robustly r

A Conditional U-Net Pipeline with Pre- and Post-Processing for Aerial RGB-to-Thermal Image Translation

arXiv:2605.17564v1 Announce Type: new Abstract: Paired RGB-thermal data has shown significant utility across a range of applications, including image fusion, object tracking, and anomaly detection; ho

A Dataset for the Recognition of Historical and Handwritten Music Scores in Western Notation

arXiv:2605.18436v1 Announce Type: new Abstract: A large amount of musical heritage has been digitised by memory institutions: libraries, museums, and archives. Nevertheless, the field of Optical Music

A Dexterous and Compliant Gripper With Soft Hydraulic Actuation for Microgravity Manipulation

arXiv:2605.17851v1 Announce Type: new Abstract: Astrobee's existing one-degree-of-freedom (DOF) underactuated compliant claw gripper enables perching on the International Space Station (ISS), but prov

researcharxiv-cs-ro

A Distributional View for Visual Mechanistic Interpretability: KL-Minimal Soft-Constraint Principle

arXiv:2605.17504v1 Announce Type: cross Abstract: Most current paradigms in visual mechanistic interpretability (MI) remain confined to interpreting internal units of the vision model via heuristic me

A Holistic Method for Superquadric Fitting Using Unsupervised Clustering Analysis

arXiv:2605.16779v1 Announce Type: cross Abstract: This work presents a novel method for fitting superquadrics to point clouds under the contamination of noise and outliers, which has many applications

A Large-Scale Study on the Accuracy vs Cost Trade-offs of Training and Evaluation Settings in Fine-Grained Image Recognition

arXiv:2605.18700v1 Announce Type: new Abstract: Prior work on fine-grained image recognition (FGIR) has established the importance of the backbone selection, but has neglected the accuracy-vs-cost tra

A More Word-like Image Tokenization for MLLMs

arXiv:2605.17954v1 Announce Type: cross Abstract: Modern multimodal large language models (MLLMs) typically keep the language model fixed and train a visual projector that maps the pixels into a seque

A neurosymbolic Approach with Epistemic Deep Learning for Hierarchical Image Classification

arXiv:2605.16383v1 Announce Type: cross Abstract: Deep neural networks achieve high accuracy on image classification tasks. Yet, they often produce overconfident predictions as which fail to express e

A New Perspective on Precision and Recall for Generative Models

arXiv:2511.02414v2 Announce Type: replace Abstract: With the recent success of generative models in image and text, the question of their evaluation has recently gained a lot of attention. While most

A No-Defense Defense Against Gradient-Based Adversarial Attacks on ML-NIDS: Is Less More?

arXiv:2605.18666v1 Announce Type: new Abstract: Gradient-based adversarial attacks subtly manipulate inputs of Machine Learning (ML) models to induce incorrect predictions. This paper investigates whe

A Novel Model for 3D Motion Planning for a Generalized Dubins Vehicle with Pitch and Yaw Rate Constraints

arXiv:2509.24143v2 Announce Type: replace Abstract: In this paper, we propose a new modeling approach and a fast algorithm for 3D motion planning, applicable for fixed-wing unmanned aerial vehicles. T

researcharxiv-cs-ro

A Practical Noise2Noise Denoising Pipeline for High-Throughput Raman Spectroscopy

arXiv:2605.18511v1 Announce Type: new Abstract: A lightweight and reproducible denoising pipeline for high-throughput Raman spectroscopy is presented. The approach relies on a one-dimensional convolut

A Readiness-Driven Runtime for Pipeline-Parallel Training under Runtime Variability

arXiv:2605.18750v1 Announce Type: cross Abstract: Pipeline parallelism is a key technique for scaling large-model training, but modern workloads exhibit runtime variability in computation and communic

A Retrieval-Augmented Generation Approach to Extracting Algorithmic Logic from Neural Networks

arXiv:2512.04329v2 Announce Type: replace Abstract: Reusing existing neural-network components is central to research efficiency, yet discovering, extracting, and validating such modules across thousa

A Scalable Tool for Measuring Manner and Result Verbs in Developmental Language Research

arXiv:2605.16654v1 Announce Type: cross Abstract: Manner and result verbs encode different aspects of event structure and have been discussed in developmental work as a potentially informative distinc

A semantic mutation metric for metamorphic relation adequacy in scientific computing programs

arXiv:2605.17437v1 Announce Type: cross Abstract: Context. Metamorphic Testing addresses the test-oracle problem in scientific computing, but classical Mutation Score operates on syntactic AST mutatio

A simple approach for biometrics: Finger-knuckle prints recognition based on a Sobel filter and similarity measures

arXiv:2605.17673v1 Announce Type: new Abstract: The objective of this work is to propose a novel methodology for the finger knuckle print recognition, which is essentially a digital photo of the finge

A Structural Threshold in Decision Capacity Governs Collapse in Self-Play Reinforcement Learning

arXiv:2605.16315v1 Announce Type: cross Abstract: We show that a threshold in decision capacity determines whether self-play reinforcement learning agents collapse under asymmetric rule perturbations.

A Systematic Survey on Deep Learning Architectures for Point Cloud Classification and Segmentation

arXiv:2605.17131v1 Announce Type: cross Abstract: Point cloud stands as the most widely adopted format for representing 3D shapes and scenes due to its simplicity and geometric fidelity. However, its

A Theory of Training Profit-Optimal LLMs

arXiv:2605.16430v1 Announce Type: cross Abstract: Scaling LLMs requires tremendous computational resources, and recent advances in AI have gone hand in hand with massive amounts of capital expenditure

A Unified Framework for Data-Free One-Step Sampling via Wasserstein Gradient Flows

arXiv:2605.17808v1 Announce Type: new Abstract: We develop a unified theoretical framework for data-free one-step sampling from unnormalized target distributions based on Wasserstein gradient flows. F

A Unified Framework for Structured Flow Modeling: From Continuous Fields to Data-Driven Representations

arXiv:2605.18250v1 Announce Type: cross Abstract: Many dynamical systems can be described in terms of structured flows combining source/sink behavior, cyclic dynamics, and topology-constrained transpo

AASIST3: KAN-Enhanced AASIST Speech Deepfake Detection using SSL Features and Additional Regularization for the ASVspoof 2024 Challenge

arXiv:2408.17352v2 Announce Type: replace-cross Abstract: Automatic Speaker Verification (ASV) systems, which identify speakers based on their voice characteristics, have numerous applications, such a

Accelerating AI-Powered Research: The PuppyChatter Framework for Usable and Flexible Tooling

arXiv:2605.17809v1 Announce Type: new Abstract: This research addresses the challenges inherent in developing Artificial Intelligence (AI) applications, particularly those leveraging Large Language Mo

Accelerating Inference for Multilayer Neural Networks with Quantum Computers

arXiv:2510.07195v2 Announce Type: replace-cross Abstract: Fault-tolerant Quantum Processing Units (QPUs) promise to deliver exponential speed-ups in select computational tasks, yet their integration i

Accelerating Redshift-Conditioned Galaxy Image Synthesis with One-step Generative Modeling

arXiv:2605.17546v1 Announce Type: cross Abstract: Understanding galaxy morphology evolution across cosmic time requires models that can generate realistic galaxy populations conditioned on redshift. I