DECO: Decoupled Multimodal Diffusion Transformer for Bimanual Dexterous Manipulation with a Plugin Tactile Adapter

Bimanual dexterous manipulation relies on integrating multimodal inputs to perform complex real-world tasks. We propose DECO, a decoupled multimodal diffusion transformer that disentangles vision, proprioception, and tactile signals through specialized conditioning pathways, with a lightweight adapter for parameter-efficient injection of additional signals. Alongside DECO, we release DECO-50 dataset for bimanual dexterous manipulation with tactile sensing, consisting of 50 hours of data and over 5M frames. Experimental results show that DECO achieves the72.25% average success rate with a 21% improvement over the baseline.

arXiv preprint