Retrospective Harness Optimization: Improving LLM Agents via Self-Preference over Trajectory Rollouts Paper • 2606.05922 • Published 24 days ago • 69
OmniDirector: General Multi-Shot Camera Cloning without Cross-Paired Data Paper • 2606.13432 • Published 17 days ago • 111
Robust-U1: Can MLLMs Self-Recover Corrupted Visual Content for Robust Understanding? Paper • 2606.08063 • Published 22 days ago • 81
Policy and World Modeling Co-Training for Language Agents Paper • 2606.02388 • Published 27 days ago • 11
AsyncTool: Evaluating the Asynchronous Function Calling Capability under Multi-Task Scenarios Paper • 2605.27995 • Published May 27 • 16
PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective Paper • 2605.28819 • Published May 27 • 8
SocialOmni: Benchmarking Audio-Visual Social Interactivity in Omni Models Paper • 2603.16859 • Published Mar 17 • 249