RoboInter: A Holistic Intermediate Representation Suite Towards Robotic Manipulation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Hao, Wang, Ziqin, Ding, Zi-han, Yang, Shuai, Chen, Yilun, Tian, Yang, Hu, Xiaolin, Wang, Tai, Lin, Dahua, Zhao, Feng, Liu, Si, Pang, Jiangmiao
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918331242512384
author Li, Hao
Wang, Ziqin
Ding, Zi-han
Yang, Shuai
Chen, Yilun
Tian, Yang
Hu, Xiaolin
Wang, Tai
Lin, Dahua
Zhao, Feng
Liu, Si
Pang, Jiangmiao
author_facet Li, Hao
Wang, Ziqin
Ding, Zi-han
Yang, Shuai
Chen, Yilun
Tian, Yang
Hu, Xiaolin
Wang, Tai
Lin, Dahua
Zhao, Feng
Liu, Si
Pang, Jiangmiao
contents Advances in large vision-language models (VLMs) have stimulated growing interest in vision-language-action (VLA) systems for robot manipulation. However, existing manipulation datasets remain costly to curate, highly embodiment-specific, and insufficient in coverage and diversity, thereby hindering the generalization of VLA models. Recent approaches attempt to mitigate these limitations via a plan-then-execute paradigm, where high-level plans (e.g., subtasks, trace) are first generated and subsequently translated into low-level actions, but they critically rely on extra intermediate supervision, which is largely absent from existing datasets. To bridge this gap, we introduce the RoboInter Manipulation Suite, a unified resource including data, benchmarks, and models of intermediate representations for manipulation. It comprises RoboInter-Tool, a lightweight GUI that enables semi-automatic annotation of diverse representations, and RoboInter-Data, a large-scale dataset containing over 230k episodes across 571 diverse scenes, which provides dense per-frame annotations over more than 10 categories of intermediate representations, substantially exceeding prior work in scale and annotation quality. Building upon this foundation, RoboInter-VQA introduces 9 spatial and 20 temporal embodied VQA categories to systematically benchmark and enhance the embodied reasoning capabilities of VLMs. Meanwhile, RoboInter-VLA offers an integrated plan-then-execute framework, supporting modular and end-to-end VLA variants that bridge high-level planning with low-level execution via intermediate supervision. In total, RoboInter establishes a practical foundation for advancing robust and generalizable robotic learning via fine-grained and diverse intermediate representations.
format Preprint
id arxiv_https___arxiv_org_abs_2602_09973
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle RoboInter: A Holistic Intermediate Representation Suite Towards Robotic Manipulation
Li, Hao
Wang, Ziqin
Ding, Zi-han
Yang, Shuai
Chen, Yilun
Tian, Yang
Hu, Xiaolin
Wang, Tai
Lin, Dahua
Zhao, Feng
Liu, Si
Pang, Jiangmiao
Robotics
Advances in large vision-language models (VLMs) have stimulated growing interest in vision-language-action (VLA) systems for robot manipulation. However, existing manipulation datasets remain costly to curate, highly embodiment-specific, and insufficient in coverage and diversity, thereby hindering the generalization of VLA models. Recent approaches attempt to mitigate these limitations via a plan-then-execute paradigm, where high-level plans (e.g., subtasks, trace) are first generated and subsequently translated into low-level actions, but they critically rely on extra intermediate supervision, which is largely absent from existing datasets. To bridge this gap, we introduce the RoboInter Manipulation Suite, a unified resource including data, benchmarks, and models of intermediate representations for manipulation. It comprises RoboInter-Tool, a lightweight GUI that enables semi-automatic annotation of diverse representations, and RoboInter-Data, a large-scale dataset containing over 230k episodes across 571 diverse scenes, which provides dense per-frame annotations over more than 10 categories of intermediate representations, substantially exceeding prior work in scale and annotation quality. Building upon this foundation, RoboInter-VQA introduces 9 spatial and 20 temporal embodied VQA categories to systematically benchmark and enhance the embodied reasoning capabilities of VLMs. Meanwhile, RoboInter-VLA offers an integrated plan-then-execute framework, supporting modular and end-to-end VLA variants that bridge high-level planning with low-level execution via intermediate supervision. In total, RoboInter establishes a practical foundation for advancing robust and generalizable robotic learning via fine-grained and diverse intermediate representations.
title RoboInter: A Holistic Intermediate Representation Suite Towards Robotic Manipulation
topic Robotics
url https://arxiv.org/abs/2602.09973