SWAM model released
The story so far
Hugging Face has released the SWAM extension method for general world action models based on cross-modal alignment. By training on 47,000 hours of multimodal data, it combines egocentric video and UMI demonstrations to improve control generalization, with the final model achieving a success rate of 40.00% in three real tasks, surpassing the baseline of 23.33%. The model release included organized data, training code, and model weights.
AI-written from the reports · Updated 7 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- Hugging Face Daily PapersSWAM: Scaling World Action Models with Cross-Modal Alignment
SWAM extends the World Action Model through cross-modal alignment, trained on 47,000 hours of multimodal data, blending egocentric video and UMI demonstrations to enhance control generalization, ultimately achieving a 40.00% success rate on three real-world tasks, surpassing the baseline of 23.33%.
Heat over time
Not enough data yet to draw a trend.