Author

Yongxin Chen

h-index 5 View on Semantic Scholar

We have 1 of 8 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Jul 2026

Understanding and Mitigating the Video-Action Generalization Gap via Temporal Ratio

An inference-time adaptive guidance method is proposed, which exploits this intrinsic feature attention pattern to dynamically amplify compositional video conditioning signals precisely when the policy relies on future rollouts, and mitigates the OOD-ID compositional generalization gap.

U. A. Mishra, Yongxin Chen, Danfei Xu et al. · arXiv.org · 1 citation