AI News

Where Does the Signal Live? A Web Data Recipe for Medical Encoder Pretraining (arxiv.org)

Rotation-free Online Handwritten Character Recognition Using Linear Recurrent Units (arxiv.org)

Observing Health Outcomes Using Remote Sensing Imagery and Geo-Context Guided Visual Transformer (arxiv.org)

High-probability guarantees for linear accessibility in feature superposition (arxiv.org)

SWE-Tester: Training Open-Source LLMs for Issue Reproduction in Real-World Repositories (arxiv.org)

Residual-augmented flow matching operators for probabilistic partial differential equations (arxiv.org)

Unexplored flaws in multiple-choice VQA make benchmarking unreliable (arxiv.org)

Latent-to-Latent Flow for Volumetric Stochastic Segmentation (arxiv.org)

Generalized infinite dimensional Alpha-Procrustes based geometries (arxiv.org)

FATS: A Prompt Injection Attack Utilizing Feign Security Agents with Deceptive Few-shots Learning (arxiv.org)

Evidential-Based Higher-Order Set Argumentation Framework (arxiv.org)

PocketAgents: A Manifest-Driven Library of Autonomous Defense Agents (arxiv.org)

WildRelight: A Real-World Benchmark and Physics-Guided Adaptation for Single-Image Relighting (arxiv.org)

Directed mixed membership stochastic blockmodel (arxiv.org)

Learn to Rank: Visual Attribution by Learning Importance Ranking (arxiv.org)

Hyperspectral Trajectory Image for Multi-Month Trajectory Anomaly Detection (arxiv.org)

MGD: Moment Guided Diffusion for Maximum Entropy Generation (arxiv.org)

Interpretable Network-assisted Random Forest+ (arxiv.org)

Differentially Private Model-X Knockoffs via Johnson-Lindenstrauss Transform (arxiv.org)

Formal Bayesian Transfer Learning via the Total Risk Prior (arxiv.org)

Introducing HALC: A general pipeline for the systematic and reliable construction of prompts for automated coding with LLMs in the computational social sciences (arxiv.org)

Mind the Gap: Navigating Inference with Optimal Transport Maps (arxiv.org)

MILAAP: Mobile Link Allocation via Attention-based Prediction (arxiv.org)

Fewer yet critical: Reducing Redundant Token Dependencies for Transformer-based Time Series Forecasting (arxiv.org)

RePro: Training Language Models to Faithfully Recycle the Web for Pretraining (arxiv.org)

Efficient and Microphone-Fault-Tolerant 3D Sound Source Localization (arxiv.org)

ControlTac: Scaling Tactile Data with Physically Controlled Tactile Image Generation (arxiv.org)

How to Backdoor Image Knowledge Distillation (arxiv.org)

DDPM Score Matching and Distribution Learning (arxiv.org)

Approaching the Harm of Gradient Attacks While Only Flipping Labels (arxiv.org)

AhaBench: Do Agents Learn from Prior Experience? A Benchmark for Long-Horizon Continual Learning (arxiv.org)

On Generalisation Error Bounds for Transformers (arxiv.org)

The EM-algorithm and the Method of Moments in Softmax Mixture Models (arxiv.org)

Tempora-Fusion: Time-Lock Puzzle with Efficient Verifiable Homomorphic Linear Combination (arxiv.org)

Decision-Centered Abstractions via Orthogonal Estimation of Difference-of-Q Functions (arxiv.org)

Decoder-Side Semantic Conditioning for Low-Bitrate Neural Speech Compression (arxiv.org)

D-ADD: An Effective Plug-In for Defending Against Model Stealing (arxiv.org)

Fine-Grained Instruction-Guided Graph Reasoning for Vision-and-Language Navigation (arxiv.org)

Precision at Scale: Domain-Specific Datasets On-Demand (arxiv.org)

Attribution in Scientific Literature: New Benchmark and Methods (arxiv.org)

Information entropy as an anthropomorphic concept (arxiv.org)

How to Build a Quantum Supercomputer: Scaling from Hundreds to Millions of Qubits (arxiv.org)

Linear and Quadratic Discriminant Analysis: Tutorial (arxiv.org)

Multi-Task Learning with Covariate-Overlap Regularization (arxiv.org)

From Proxies to Fields: Spatiotemporal Reconstruction of Global Radiation from Sparse Sensor Sequences (arxiv.org)

Evaluating the Scalability and Adversarial Generalization of GRPO-Trained NLI Models (arxiv.org)

Deep Learning Approach to Bearing and Induction Motor Fault Diagnosis via Data Fusion (arxiv.org)

NoLoCo: No-all-reduce Low Communication Training Method for Large Models (arxiv.org)

Two-dimensional Taxonomy for N-ary Knowledge Representation Learning Methods (arxiv.org)

A Theoretical Analysis of Provable Compositional Generalization in Neural Networks: A Necessary and Sufficient Condition (arxiv.org)