AI Paper Atlas

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

AtomicVLA: Unlocking the Potential of Atomic Skill Learning in Robots

Likui Zhang, Tao Tang, Zhihao Zhan, Xiuwei Chen, Zisheng Chen, Jianhua Han, Jiangtong Zhu, Pei Xu, Hang Xu, Hefeng Wu, Liang Lin, Xiaodan Liang

CVPR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 19 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
add skill-guided MoE (SG-MoE) routing on atomic skills (ablation)pi0 · LIBERO-Longsuccess rate (%): 85.2 -> 95.2Improved
add generic token-level MoE (ablation)pi0 · LIBERO-Longsuccess rate (%): 85.2 -> 88.6Improved
add timestep-conditioned MoDE routing (ablation)pi0 · LIBERO-Longsuccess rate (%): 85.2 -> 89.5Improved
route experts by atomic skill instead of token-level MoE (ablation)AtomicVLA · LIBERO-Longsuccess rate (%): 88.6 -> 95.2Improved
route experts by atomic skill instead of timestep (MoDE) (ablation)AtomicVLA · LIBERO-Longsuccess rate (%): 89.5 -> 95.2Improved
add atomic-skill MoE and planning on pi0pi0 · LIBERO averagesuccess rate (%): 94.2 -> 96.6Improved
add atomic-skill MoE and planning on pi0.5pi0.5 · LIBERO averagesuccess rate (%): 96.9 -> 97.8Improved
add atomic-skill MoE on pi0pi0 · CALVIN ABC->Daverage length: 3.87 -> 4.09Improved
add atomic-skill MoE on pi0.5pi0.5 · CALVIN ABC->Daverage length: 4.02 -> 4.27Improved
add atomic-skill MoE for multi-task long-horizon trainingpi0.5 · Real-world long-horizon taskssuccess rate (%) average: 45 -> 63.3Improved
learn new skill by adding an expert instead of fine-tuning the single decoderpi0.5 · Real-world continual learningsuccess rate (%) average on base tasks: -15.0 -> -1.3Improved
add atomic-skill MoEpi0.5 · Real-world complex scenessuccess rate (%) average: 33.3 -> 43.3Improved

Later papers that compared with this method

2 later accepted papers report AtomicVLA as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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