Continuous Vision-Language-Action Co-Learning with Semantic-Physical Alignment for Behavioral Cloning

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Qi, Xiuxiu, Yang, Yu, Cao, Jiannong, Bai, Luyao, Fan, Chongshan, Cao, Chengtai, Wang, Hongpeng
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908729508626432
author Qi, Xiuxiu
Yang, Yu
Cao, Jiannong
Bai, Luyao
Fan, Chongshan
Cao, Chengtai
Wang, Hongpeng
author_facet Qi, Xiuxiu
Yang, Yu
Cao, Jiannong
Bai, Luyao
Fan, Chongshan
Cao, Chengtai
Wang, Hongpeng
contents Language-conditioned manipulation facilitates human-robot interaction via behavioral cloning (BC), which learns control policies from human demonstrations and serves as a cornerstone of embodied AI. Overcoming compounding errors in sequential action decisions remains a central challenge to improving BC performance. Existing approaches mitigate compounding errors through data augmentation, expressive representation, or temporal abstraction. However, they suffer from physical discontinuities and semantic-physical misalignment, leading to inaccurate action cloning and intermittent execution. In this paper, we present Continuous vision-language-action Co-Learning with Semantic-Physical Alignment (CCoL), a novel BC framework that ensures temporally consistent execution and fine-grained semantic grounding. It generates robust and smooth action execution trajectories through continuous co-learning across vision, language, and proprioceptive inputs (e.g., robot internal states). Meanwhile, we anchor language semantics to visuomotor representations by a bidirectional cross-attention to learn contextual information for action generation, successfully overcoming the problem of semantic-physical misalignment. Extensive experiments show that CCoL achieves an average 8.0% relative improvement across three simulation suites, with up to 19.2% relative gain in human-demonstrated bimanual insertion tasks. Real-world tests on a 7-DoF robot further confirm CCoL's generalization under unseen and noisy object states.
format Preprint
id arxiv_https___arxiv_org_abs_2511_14396
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Continuous Vision-Language-Action Co-Learning with Semantic-Physical Alignment for Behavioral Cloning
Qi, Xiuxiu
Yang, Yu
Cao, Jiannong
Bai, Luyao
Fan, Chongshan
Cao, Chengtai
Wang, Hongpeng
Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
Language-conditioned manipulation facilitates human-robot interaction via behavioral cloning (BC), which learns control policies from human demonstrations and serves as a cornerstone of embodied AI. Overcoming compounding errors in sequential action decisions remains a central challenge to improving BC performance. Existing approaches mitigate compounding errors through data augmentation, expressive representation, or temporal abstraction. However, they suffer from physical discontinuities and semantic-physical misalignment, leading to inaccurate action cloning and intermittent execution. In this paper, we present Continuous vision-language-action Co-Learning with Semantic-Physical Alignment (CCoL), a novel BC framework that ensures temporally consistent execution and fine-grained semantic grounding. It generates robust and smooth action execution trajectories through continuous co-learning across vision, language, and proprioceptive inputs (e.g., robot internal states). Meanwhile, we anchor language semantics to visuomotor representations by a bidirectional cross-attention to learn contextual information for action generation, successfully overcoming the problem of semantic-physical misalignment. Extensive experiments show that CCoL achieves an average 8.0% relative improvement across three simulation suites, with up to 19.2% relative gain in human-demonstrated bimanual insertion tasks. Real-world tests on a 7-DoF robot further confirm CCoL's generalization under unseen and noisy object states.
title Continuous Vision-Language-Action Co-Learning with Semantic-Physical Alignment for Behavioral Cloning
topic Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.14396