ChunkFlow: Towards Continuity-Consistent Chunked Policy Learning
- Published
- Source
- arXiv
- Paper number
- 628
- Field
- Robotics
- arXiv ID
- 2607.12992
Key points
- It divides each action chunk into fixed, revisable, and future regions and deterministically blends the overlapping portions during execution.
- It trains with seam loss and first- and second-order continuity losses, using history corruption, scheduled sampling, and AWAC fine-tuning to improve robustness to execution errors.
- In CALVIN, LIBERO, and real-robot experiments, it improved the balance between success rate and motion stability while maintaining low inference latency.
- By reducing jitter at chunk boundaries during training rather than through post-processing filters, it can be applied to real-time VLA robot control.
- The theoretical result that seam discrepancies decrease with overlap length holds under mild smoothness assumptions, and the experiments are limited to three environments.
Paper links
External research summaries. These are not HDATF publications or measured product results.