Heave-Ho framework released
The story so far
Hugging Face has released the decentralized long-term multi-humanoid collaboration framework Heave-Ho, which achieves long-term collaboration of humanoid robots through low-level multi-agent post-training and advanced visual language planning, employing a voice-based leader-follower synchronization mechanism that requires no network communication. The system, trained with two identical humanoid robots, can be scaled up to a maximum of 10 robots and generalizes without retraining in heterogeneous teams, with HH-Bench evaluations showing its continuous superiority over baseline methods.
AI-written from the reports · Updated 6 hours ago
Coverage timeline
Follow the coverage to see every side of the story.
- Hugging Face Daily PapersHeave-Ho: Decentralized Long-Horizon Multi-Humanoid Collaboration
The Heave-Ho framework achieves long-term collaboration of humanoid robots through low-level multi-agent post-training and advanced visual language planning, employing a voice-based leader-follower synchronization mechanism that requires no network communication. The system, trained with two identical humanoid robots, can be scaled up to a maximum of 10 robots and generalizes without retraining in heterogeneous teams, with HH-Bench evaluations showing its continuous superiority over baseline methods.
Heat over time
- Now
- 6
- Peak
- 619 Nov, 00:00
- 24h change
- –
Hover or tap the chart to see each hour; use the arrow keys on a keyboard.