Event
PLaCA: A Robot Action Generation Framework for Predicting Future Latent Variable Changes
1 reports · 1 sources
Zhan Yi and other researchers propose PLaCA, a world-to-action framework that directly predicts spatially structured patch-level future latent variable changes from current multimodal input and connects them to action generation. On LIBERO-Long, PLaCA achieves a success rate of 79.0%, a 5.0 percentage point improvement over the Action-only baseline; on CALVIN, the average sequence length improves from 2.784 to 3.564. The second-stage training updates only 2.95% of the model parameters.
Report timeline
- PLaCA: A Robot Action Generation Framework Based on Future Latent Change
PLaCA proposes a world-to-action framework that directly predicts spatially structured patch-level future latent change from current multimodal input and explicitly incorporates it into action generation. On LIBERO-Long, PLaCA achieves a 79.0% success rate, a 5.0 percentage point improvement over the Action-only baseline; on CALVIN, the average sequence length improves from 2.784 to 3.564. The second-stage training updates only 2.95% of the model parameters.
arXiv cs.RO ·