Research map / Vision-language and multimodal / Robot action and navigation
ChatVLA-2: Vision-Language-Action Model with Open-World Reasoning
Zhongyi Zhou, Yichen Zhu, Xiaoyu Liu, Zhibin Tang, Junjie Wen, Yaxin Peng, Chaomin Shen, Yi Xu
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes ChatVLA-2, a vision-language-action model that aims to retain and extend VLM capabilities such as open-world reasoning while performing robotic manipulation.
Problem
End-to-end vision-language-action models often lose key pre-trained VLM capabilities (such as open-world reasoning) during fine-tuning on specific robotic tasks.
Approach
Argues a generalizable VLA should retain and expand the VLM's core competencies, including open-world reasoning, rather than sacrificing them during robot fine-tuning; the abstract does not detail the mechanism.
Closest papers by meaning
- CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models (CVPR 2025 · similarity 0.92)
- ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models (NeurIPS 2026 · similarity 0.91)
- Seeing Across Views: Benchmarking Spatial Reasoning of Vision-Language Models in Robotic Scenes (ICLR 2026 · similarity 0.91)
- Knowledge Insulating Vision-Language-Action Models: Train Fast, Run Fast, Generalize Better (NeurIPS 2025 · similarity 0.90)
- UP-VLA: A Unified Understanding and Prediction Model for Embodied Agent (ICML 2025 · similarity 0.90)
- PIVOT: Iterative Visual Prompting Elicits Actionable Knowledge for VLMs (ICML 2024 · similarity 0.90)
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.
