*.arxiv.org

21806 posts in the last 30 days

OffsetAxis: UDF Mesh Reconstruction via Offset-Volume Medial Axis Extraction

Evaluating Chinese Ambiguity Understanding in Large Language Models

PRISM: Prompt Reliability via Iterative Simulation and Monitoring for Enterprise Conversational AI

$f$-Trajectory Balance: A Loss Family for Tuning GFlowNets, Generative Models, and LLMs with Off- and On-Policy Data

Second-Order Multi-Level Variance Correction for Modality Competition in Multimodal Models

Joint Mobile User Positioning and Passive Target Sensing using Optimized Sequential Beamforming

GQA-{\mu}P: The maximal parameterization update for grouped query attention

Entropic Auto-Encoding via Implicit Free-Energy Minimization

CM-EVS: Sparse Panoramic RGB-D-Pose Data for Complete Scene Coverage

MI-CXR: A Benchmark for Longitudinal Reasoning over Multi-Interval Chest X-rays

Process Rewards with Learned Reliability

LDGuid: A Framework for Robust Change Detection via Latent Difference Guidance

Verifiers and Generators: Epistemic Semantics for Intuitionistic Logic (Long Version)

Method of Fundamental Solutions for Maxwell's Equations in Bi-Periodic Multilayered Media

Ada-Diffuser: Latent-Aware Adaptive Diffusion for Decision-Making

Sustainability in Telecom: Energy-Efficient Networks and Circular Economy Models to Reduce Carbon Footprints and Increase Efficiency

Method-level Change-proneness: A Better Metric for Black-box Test Suite Minimization

Statistical two-round search for one excellent element

FRWKV+: Adaptive Periodic-Position Branch Interaction for Frequency-Space Linear Time Series Forecasting

Interpreting De Finetti's theorem in the Category of Integrable Cones (long version)

Log in to enable infinite scrolling