AI Paper Atlas

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

Action-aware Dynamic Pruning for Efficient Vision-Language-Action Manipulation

Xiaohuan Pei, Yuxing Chen, Siyu Xu, Yunke Wang, Yuheng Shi, Chang Xu

ICLR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models

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
add action-aware dynamic gating to text-driven pruning (ablation)VLA-ADP on OpenVLA-OFT · LIBERO averagesuccess rate: 93.45% -> 96.3%Improved
use dynamic gating instead of periodic switching (ablation)VLA-ADP on OpenVLA-OFT · LIBERO-Objectsuccess rate: 81.4% -> 98.0%Improved
use dynamic gating instead of periodic switching (ablation)VLA-ADP on OpenVLA-OFT · LIBERO averagesuccess rate: 89.9% -> 96.3%Improved
score tokens at deeper layers (1 or 4) instead of layer 0 (ablation)VLA-ADP on OpenVLA-OFT · LIBERO averagesuccess rate: 96.3% -> 95.5% (layer 1), 95.8% (layer 4)Hurt
score tokens at layer 4 instead of layer 0 (ablation)VLA-ADP on OpenVLA-OFT · LIBERO averageFLOPs: 6.43 -> 6.89Hurt
text-driven relevance pruning instead of random token dropping (50%)OpenVLA-OFT · LIBERO-Objectsuccess rate: 73.0% -> 98.0%Improved
apply ADP at keep ratio 50-70% to OpenVLA-OFTOpenVLA-OFT · LIBERO averagesuccess rate: 97.1% -> 96.3% with 1.23x speedupMixed
apply ADP instead of FastV pruning on OpenVLA-OFTOpenVLA-OFT · LIBERO averagesuccess rate: 86.8% -> 96.3%Improved
apply ADP to OpenVLA-OFT on real robotOpenVLA-OFT · Real-world (Jaco2, 4 tasks)average success rate: 85.8% -> 88.3%, latency 76.9 -> 51.8Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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