AI Paper Atlas

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

Self-Improving Vision-Language-Action Models with Data Generation via Residual RL

Wenli Xiao, Haotian Lin, Andy Peng, Haoru Xue, Tairan He, Zhengyi Luo, Yuqi Xie, Fengyuan Hu, Linxi Fan, Guanya Shi, Yuke Zhu

ICLR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 62 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
fine-tune on PLD data in addition to SFTpi0 · LIBERO averagesuccess rate (%): 93.4 -> 97.2Improved
fine-tune on PLD data in addition to OFTOpenVLA · LIBERO averagesuccess rate (%): 91.8 -> 99.2Improved
fine-tune on PLD data in addition to OFTOpenVLA · LIBERO-Objectsuccess rate (%): 99.1 -> 99.1No clear effect
fine-tune on PLD dataOcto · SimplerEnv averagesuccess rate (%): 71.8 -> 96.6Improved
train on PLD data instead of base policy rollout datapi0 · LIBERO-90 (full coverage)success rate: 0.488 -> 0.871Improved
train on PLD data instead of human datapi0 · LIBERO-90 (full coverage)success rate: 0.815 -> 0.871Improved
train on PLD data instead of human datapi0 · LIBERO-90 (0.1 coverage)success rate: 0.272 -> 0.314Improved
use PLD data instead of human datapi0 · Real-world blue cube pick-upsuccess rate (%): 40.0 -> 93.3Improved
use PLD data instead of human datapi0 · Real-world red cube pick-upsuccess rate (%): 33.3 -> 66.7Improved
use PLD data instead of human datapi0 · Real-world peg insertionsuccess rate (%): 100.0 -> 100.0No clear effect
increase base policy probing horizon alpha (ablation)PLD · LIBERO-90 tasks 0-9success rateMixed
scale PLD data from 50 to 500 trajectoriespi0 · LIBERO-90 few-shot from LIBERO-Goalsuccess rateImproved

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