Unverified50% confidenceFactExact time
VLA将机器人的物理动作表示为动作token，在统一序列预测框架中端到端完成从感知到行动的推理，Google DeepMind的RT-2系列工作是这一范式的代表
1
Sources
50%
Confidence
Long-term
Relevance
7/18/2026
First Seen
Sources
Related Claims
Unverified基于视觉语言模型（VLM）的屏幕理解使Agent能以语义层面而非像素层面理解界面，这一范式转变被业界称为认知自动化（Cognitive Automation）69% similarUnverifiedKrishna认为机器人模型应直接从视觉与传感器输入映射到动作，而非在中间嵌入LLM68% similarUnverifiedIn the era of large models, within an Agent architecture, the LLM acts as the brain responsible for reasoning and decision-making, tools act as the limbs responsible for interacting with the external world, and the memory module stores conversation history and intermediate states.66% similarUnverified该项目中roboreg负责机器人位姿估计，OpTaS负责实时求解带约束的导纳任务65% similarUnverifiedAndrew Ng's team is exploring the direction of embodied intelligence that combines large-model-driven agents with robotic hardware.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/552106API
curl https://kongchang.com/api/v1/knowledge/claims/552106MCP
get_claim(id=552106)