AI Paper Atlas

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

From Seeing to Doing: Bridging Reasoning and Decision for Robotic Manipulation

Yifu Yuan, Haiqin Cui, Yibin Chen, Zibin Dong, Fei Ni, Longxin Kou, Jinyi Liu, Pengyi Li, YAN ZHENG, Jianye HAO

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 SrCoT spatial-relationship chain-of-thought reasoning (ablation)FSD · VABench-Pointaccuracy: 26.21 -> 61.82Improved
add self-consistency alignment (inverse understanding task) (ablation)FSD · VABench-Pointaccuracy: 55.92 -> 61.82Improved
add SrCoT reasoning (ablation)FSD · VABench-VisualTraceRMSE: 99.53 -> 78.26Improved
add SrCoT reasoning (ablation)FSD · VABench-VisualTraceLLM Score: 5.07 -> 6.21Improved
add self-consistency alignment (ablation)FSD · VABench-VisualTraceRMSE: 80.48 -> 78.26Improved
add self-consistency alignment (ablation)FSD · VABench-VisualTraceLLM Score: 5.92 -> 6.21Improved
train on weak-to-strong spatial data on top of ASMv2ASMv2-13B · EmbSpatial-Bench Testaccuracy: 57.4 -> 63.3Improved
use SrCoT-based FSD instead of RoboPoint for free-space referenceFSD vs RoboPoint · Where2Placeaccuracy: 46.0 vs 45.8No clear effect
reasoning-based FSD instead of end-to-end DINOv2 Predictor trained on same trace datavisual trace prediction · VABench-VisualTraceRMSE: 128.32 -> 78.26Improved
zero-shot FSD vs RoboPoint zero-shotWidowX robot · SimplerEnv WidowXaverage success rate (%): 17.7 -> 40.6Improved
zero-shot FSD vs fine-tuned pi0-fastWidowX robot · SimplerEnv WidowXaverage success rate (%): 48.3 vs 40.6Hurt

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