AI Paper Atlas

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

From Abstraction to Instantiation: Learning Behavioral Representation for Vision-Language-Action Model

Bing Hu, Zaijing Li, Rui Shao, Junda Chen, April Hua Liu, Wei-Shi Zheng, Liqiang Nie

ICML 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 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
add VBE and PBD on top of pi0.5 (ablation)pi0.5 · Real-world Generalizationsuccess rate: 57.0 -> 70.0Improved
add VBE and PBD on top of pi0.5 (ablation)pi0.5 · Real-world Long-horizonsuccess rate: 41.0 -> 55.0Improved
add VBE and PBD on top of pi0.5 (ablation)pi0.5 · LIBERO Longsuccess rate: 92.4 -> 94.6Improved
add VBE and PBD on top of pi0.5 (ablation)pi0.5 · LIBERO Spatialsuccess rate: 98.8 -> 99.2Improved
add VBE alone or PBD alone (ablation)pi0.5 · LIBERO Spatialsuccess rate: 98.8 -> 98.2 (one single-module row)No clear effect
add VBE and PBD to pi0.5 on RoboTwin 2.0 Hardpi0.5 · RoboTwin 2.0 Hard (20 tasks avg)success rate: 49.2 -> 58.0Improved
add VBE and PBD to pi0.5 on LIBEROpi0.5 · LIBERO averagesuccess rate: 96.9 -> 98.0Improved
tune prior guidance strength lambda away from its optimum (ablation)BehaviorVLA · LIBERO suitessuccess rateHurt
retrieve k=5 prototypes instead of 1, 3 or 10 (ablation)BehaviorVLA · LIBERO Longsuccess rate: k=1 93.0, k=3 93.8, k=5 94.6, k=10 94.0Mixed
remove vision stream or action stream from VBE (ablation)VBE · t-SNE of 20 task prototypescluster separability (qualitative)Hurt
train with only 50% of real-world demonstrationsBehaviorVLA · Real-world taskssuccess rateNo clear effect

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