DAM-VLA: Decoupled Asynchronous Multimodal Vision Language Action model
This paper presents DAM-VLA, a vision-language-action model that decouples temporal processing per modality to improve performance in contact-rich real-world manipulation tasks.
DAM-VLA introduces a novel approach to decoupling temporal processing per modality, which is different from the synchronous clock used in traditional vision-language pretraining.
Before reading this…
Applications
- →Contact-rich real-world manipulation tasks
To understand this paper, make sure you know these concepts first:
- Vision-language pretrainingfind papers →
- Temporal processingfind papers →
Abstract
More Like ThisVision-language-action (VLA) models inherit a shared synchronous clock from vision-language pretraining, processing every input at one rate. This is misaligned with physical interaction, where a high-frequency modality changes at hundreds of hertz, vision evolves more slowly, and language stays constant across an episode. A synchronous VLA oversamples slow modalities, undersamples fast ones, and caps action generation at the lowest effective frequency. We hypothesize that decoupling temporal processing per modality, letting each update and retain information at its own sensor rate, yields stronger representations and more robust control. We present DAM-VLA, which maintains per-modality latent buffers refreshed at sensor rates and read continuously by the action head, integrating new high-frequency modalities through gated cross-attention that leaves the pretrained backbone intact. Across seven contact-rich real-world manipulation tasks, DAM-VLA more than doubles the average success rate of the strongest synchronous baseline (95.2\% vs.\ 40.95\%) while sustaining smooth, reactive 100\,Hz control. Project website: \href{https://intuitive-robots.github.io/DAM-VLA/}{intuitive-robots.github.io/DAM-VLA/}