Thinker: A vision-language foundation model for embodied intelligence

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pan, Baiyu, Luo, Daqin, Yang, Junpeng, Wang, Jiyuan, Zhang, Yixuan, Shi, Hailin, Jiao, Jichao
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917230827012096
author Pan, Baiyu
Luo, Daqin
Yang, Junpeng
Wang, Jiyuan
Zhang, Yixuan
Shi, Hailin
Jiao, Jichao
author_facet Pan, Baiyu
Luo, Daqin
Yang, Junpeng
Wang, Jiyuan
Zhang, Yixuan
Shi, Hailin
Jiao, Jichao
contents When large vision-language models are applied to the field of robotics, they encounter problems that are simple for humans yet error-prone for models. Such issues include confusion between third-person and first-person perspectives and a tendency to overlook information in video endings during temporal reasoning. To address these challenges, we propose Thinker, a large vision-language foundation model designed for embodied intelligence. We tackle the aforementioned issues from two perspectives. Firstly, we construct a large-scale dataset tailored for robotic perception and reasoning, encompassing ego-view videos, visual grounding, spatial understanding, and chain-of-thought data. Secondly, we introduce a simple yet effective approach that substantially enhances the model's capacity for video comprehension by jointly incorporating key frames and full video sequences as inputs. Our model achieves state-of-the-art results on two of the most commonly used benchmark datasets in the field of task planning.
format Preprint
id arxiv_https___arxiv_org_abs_2601_21199
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Thinker: A vision-language foundation model for embodied intelligence
Pan, Baiyu
Luo, Daqin
Yang, Junpeng
Wang, Jiyuan
Zhang, Yixuan
Shi, Hailin
Jiao, Jichao
Computer Vision and Pattern Recognition
Artificial Intelligence
When large vision-language models are applied to the field of robotics, they encounter problems that are simple for humans yet error-prone for models. Such issues include confusion between third-person and first-person perspectives and a tendency to overlook information in video endings during temporal reasoning. To address these challenges, we propose Thinker, a large vision-language foundation model designed for embodied intelligence. We tackle the aforementioned issues from two perspectives. Firstly, we construct a large-scale dataset tailored for robotic perception and reasoning, encompassing ego-view videos, visual grounding, spatial understanding, and chain-of-thought data. Secondly, we introduce a simple yet effective approach that substantially enhances the model's capacity for video comprehension by jointly incorporating key frames and full video sequences as inputs. Our model achieves state-of-the-art results on two of the most commonly used benchmark datasets in the field of task planning.
title Thinker: A vision-language foundation model for embodied intelligence
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2601.21199