τ₀-VLA: a Hierarchical Robot Foundation Model with World-Model-Guided Test-Time Computation

Shanghai Innovation Institute

Research official + media 3 src. ~1 min

Hierarchical robot foundation model that treats high-level subtask generation as a compute-scalable inference problem solved via world-model-guided beam search with execution memory; trained on 40,115 hours of real-world multimodal data and shows large closed-loop gains on long-horizon manipulation under distribution shift.

Why it matters

HF Daily Paper with 519 upvotes (highest of the day); large real-data hierarchical VLA with test-time compute scaling

Importance: 3/5

HF Daily Paper with 519 upvotes

robotics world-models hierarchical test-time-compute vision-language-action vla

Sources

official τ₀-VLA: a Hierarchical Robot Foundation Model with World-Model-Guided Test-Time Computation

official HF Daily Paper page for τ₀-VLA

media Model card: sii-research/tau-0-vla