Unverified60% confidenceFactExact time
Google DeepMind的RT-2首次将视觉-语言模型直接用于生成机器人动作指令
2
Sources
60%
Confidence
Long-term
Relevance
7/10/2026
First Seen
Sources
Mistral Releases Robostral Navigate: A Deep Dive into the Robot Navigation Foundation Model
hackernewshackernews7/8/2026
Related Claims
UnverifiedDeepMind launched the Auto-RT project, combining large language models with the robot control model RT-2, using the LLM as a task planning layer to decompose natural language instructions into low-level action sequences.84% similarUnverifiedDeepMind developed the VLA (Vision-Language-Action model) system, which treats the robot's physical actions on equal footing with visual tokens and language tokens.79% similarUnverifiedGoogle DeepMind's SIMA project trained a general-purpose agent capable of executing natural language instructions across multiple 3D games.77% similarUnverifiedGoogle DeepMind continues to invest in GUI understanding and interaction for multimodal models, including training models on large-scale webpage screenshot data and exploring Android device control capabilities.73% similarUnverifiedGoogle DeepMind's Genie project attempts to learn interactive world models from unlabeled video.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/460978API
curl https://kongchang.com/api/v1/knowledge/claims/460978MCP
get_claim(id=460978)