AI Paper Atlas

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

RoboInter: A Holistic Intermediate Representation Suite Towards Robotic Manipulation

Hao Li, Ziqin Wang, Zi-han Ding, Shuai Yang, Yilun Chen, Yang Tian, Xiaolin Hu, Tai Wang, Dahua Lin, Feng Zhao, Si Liu, Jiangmiao Pang

ICLR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 17 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 pretrained VLM planner (implicit intermediate, IC-E2E)Vanilla executor · Open-loop In-the-WildmOLS: 0.3086 -> 0.3218Improved
make intermediate representations explicit (EC-E2E)IC-E2E · Open-loop In-the-WildmOLS: 0.3218 -> 0.3340Improved
decouple planning and execution (Te-Modular)EC-E2E · Open-loop In-the-WildmOLS: 0.3340 -> 0.3543Improved
use visual prompts instead of textual F-CoT (Im-Modular) (ablation)Te-Modular · Open-loop In-the-WildmOLS: 0.3543 -> 0.3439Hurt
add Subtask (ablation)Vanilla · Open-loop Oracle+ExecutormOLS: 0.3086 -> 0.3146Improved
add Primitive Skill on top of Subtask (ablation)Vanilla · Open-loop Oracle+ExecutormOLS: 0.3146 -> 0.3159Improved
add Object Box (ablation)Subtask+Skill · Open-loop Oracle+ExecutormOLS: 0.3159 -> 0.3289Improved
add Trace (ablation)all other representations · Open-loop Oracle+ExecutormOLS: 0.3435 -> 0.3861Improved
add Affordance alone (ablation)Vanilla · Open-loopmOLS: 0.3086 -> 0.3322Improved
add Primitive skill alone (ablation)Vanilla · Open-loopOLS@0.01: 0.0189 -> 0.0156Hurt
initialize from Planner (IC-E2E)Vanilla · Real-world OODsuccess rate: 38.3% -> 58.3%Improved
jointly generate explicit CoT (EC-E2E)IC-E2E · Real-world IDsuccess rate: 77.3% -> 68.3%Hurt

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