AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

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

Xukun Li, Yu Sun, Lei Zhang, Bo-Sheng Huang, Yibo Peng, Yuan Meng, Haojun Jiang, Shaoxuan Xie, Guocai Yao, Alois Knoll, Zhenshan Bing, Xinlong Wang and 1 more

ICML 2026 · Vision-language and multimodal › Robot action and navigation · approach: Memory and subgoal-based manipulation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
decoupled DiT conditioning (joint self-attention for vision, AdaLN for proprioception) instead of Diffusion Policy FiLM conditioningDECO vs DP · DECO-50 all tasks (vision only)success rate (%): 51.25% -> 72.25%Improved
add the plugin tactile adapter to pretrained vision-based DECODECO · DECO-50 all taskssuccess rate (%): 72.25% -> 82.50%Improved
add the tactile adapter on contact-rich tasksDECO · Waste Disposal and Assemblysuccess rate (%): 53.13% -> 73.13%Improved
add tactile via adapter on weakly tactile-relevant tasksDECO · Pick and Placesuccesses out of 120: 103/120 -> 108/120No clear effect
concatenate tactile into the input of Diffusion Policy trained from scratchDP · Assemblysuccesses out of 80: 15/80 -> 15/80No clear effect
concatenate tactile into DP inputDP · Waste Disposalsuccesses out of 80: 30/80 -> 38/80Improved
couple tactile with proprioception conditioning (add tactile embeddings to AdaLN) instead of decoupled cross-attention (ablation)DECO from scratch · Assembly ablation stage 3successes out of 40: 29/40 -> 21/40Hurt
add tactile via decoupled cross-attention trained from scratch (ablation)DECO · Assembly ablation stage 3successes out of 40: 18/40 -> 29/40Improved
use plugin adapter (7.97M trainable params) instead of training from scratch (89.41M) (ablation)DECO tactile model · Assembly ablation stage 3successes out of 40: 29/40 -> 29/40No clear effect
swap pathways: proprioception via cross-attention and tactile via AdaLN (ablation)DECO.ds · Assembly ablation stage 3successes out of 40: 29/40 -> 0/40Hurt
swap pathways: proprioception via cross-attention and tactile via AdaLN (ablation)DECO.ds · Assembly ablation stage 2successes out of 40: 34/40 -> 17/40Hurt
add tactile (adapter or concatenation) on small smooth objects in sortingDECO and DP · Material Sorting, object 00081-p-2xsuccesses out of 20: DECO 10/20 -> DECO.p 13/20; DP 5/20 -> DP.t 12/20Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy