Shuo Wang

Low-rank adaptation in vision-language alignment not only cuts costs but also boosts performance, revealing a surprising shift from hallucination to conservatism in model behavior.

Navigation success alone is a poor predictor of real-world web agent effectiveness, as revealed by the extensive evaluation of WebRetriever's diverse protocols.

CheckRLM cuts error accumulation in reasoning chains by correcting factual inaccuracies in real-time, outperforming traditional approaches.

Agents trained with T2RD can generalize learned policies across environments without overfitting to irrelevant features, achieving state-of-the-art performance in VRL tasks.

Local motion representations can drastically improve reinforcement learning efficiency and transferability across diverse tasks, challenging the conventional global modeling approach.

Temporal correlations in video data can unlock a new level of sample efficiency and performance in Reinforcement Learning pre-training.

CAT enables Large Reasoning Models to intelligently balance efficiency and accuracy, compressing confident responses while thoroughly processing uncertain queries.

Humanoid robots can complete laboratory tasks but often fail to meet the precision required for scientific validity, exposing a critical gap in current automation efforts.