Preprint
Jul 2026
Look Before You Leap: Distilling Tree Search into Action Evaluation for Frozen VLA Models
By decoupling action proposal from consequence evaluation, SVA preserves the generalization capacity of the VLA backbone while substantially improving task success rates, and shows that SVA consistently improves generalization on unseen tasks and exhibits strong test-time scaling behavior.
Xinyi Xie, Zican Hu, Zhanyun Liu et al.
· 0 citations