Xi Li

Tree search on on-policy trajectories dramatically boosts policy gradient learning by uncovering rare high-return paths, outperforming PPO on Atari and MuJoCo with minimal bias increase.

Self-generated driving failures are an untapped goldmine: treating simulator-diagnosed trajectory repairs as a training-only scaffold improves a 4B VLA by 4.0 PDMS without incurring any inference-time compute penalties.

This work studies a route-conditioned order of unavoidable stages that every successful executor must traverse, recoverable from offline trajectories and belonging to none of them: a route-conditioned order of unavoidable stages that every successful executor must traverse.

Beacon slashes design costs by up to 93.5% in multi-chiplet accelerators by transforming detailed execution reports into actionable insights through LLM-driven agents.

V-Link bridges the gap between visual perception and action control, boosting robotic manipulation success rates by up to 31.2% through enhanced feature integration.