AI Paper Atlas

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

ReconVLA: Reconstructive Vision-Language-Action Model as Effective Robot Perceiver

Wenxuan Song, Ziyang Zhou, Han Zhao, Jiayi Chen, Pengxiang Ding, Haodong Yan, Yuxin Huang, Feilong Tang, Donglin Wang, Haoang Li

AAAI 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 61 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 gaze-region reconstruction with pretraining (implicit grounding) to the VLA baselineLLaVA-7b VLA · CALVIN ABC→DAvg. Len: 3.36 -> 3.95Improved
add explicit grounding (cropped image as extra input) to the baselinebaseline VLA · CALVIN ABC→DAvg. Len: 3.36 -> 3.61Improved
add chain-of-thought bounding box output before actionsbaseline VLA · CALVIN ABC→DAvg. Len: 3.36 -> 0.63Hurt
add large-scale visual reconstruction pretraining (ablation)ReconVLA · CALVIN ABC→DAvg. Len: 3.85 -> 3.95Improved
reconstruct only the gaze region instead of the entire image (ablation)ReconVLA (no pretraining) · CALVIN ABC→DAvg. Len: 3.42 -> 3.85Improved
add whole-image reconstruction auxiliary loss (ablation)baseline VLA (no pretraining) · CALVIN ABC→DAvg. Len: 3.36 -> 3.42Improved
add gaze-region reconstruction without pretraining (ablation)baseline VLA · CALVIN ABC→Dsuccess rate 5/5 (%): 49.0 -> 58.2Improved
add gazing mechanism on stack block taskbaseline VLA · CALVIN ABC→D, stack blocksuccess rate (%): 59.3% -> 79.5%Improved
ReconVLA vs strongest generative/VLA baselinesReconVLA · CALVIN ABCD→DAvg. Len: 4.21 -> 4.23Improved
ReconVLA on unseen real-world objects vs OpenVLA and PD-VLAReconVLA · real-world unseen taskssuccess rate: baselines nearly 0%Improved

Later papers that compared with this method

2 later accepted papers report ReconVLA 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 2 of 2 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