Show HN: VLDA-learn compact Deltas as actionable transition interfaces for robot

By nullnonenilNULL · 2026-10-22 · 1 points · 0 comments

https://baidu-baige.github.io/LoongForge/projects/vlda.html

Introducing LoongForge-VLDA, a Vision-Language-Delta-Action model trained with LoongForge: 98.5% on LIBERO, 84.5% zero-shot on LIBERO-Plus, and +17–18 points over π0.5 on real robots

Open the full discussion on BetterNews