Unverified50% confidenceFactExact time
谷歌的RT-2和DeepMind的Gato项目在探索视觉-语言-行动模型
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/8/2026
First Seen
Valid until: 12/7/2026
Sources
Related Entities
Related Claims
UnverifiedGoogle DeepMind的RT-2（2023年）将机器人动作词元化为离散token，附加到视觉-语言模型PaLI-X和PaLM-E的输出词汇表中，展现出零样本泛化能力77% similarUnverified谷歌的RT-2证明了视觉-语言模型可以直接输出机器人动作指令76% similarUnverifiedGoogle DeepMind推动了A2A（Agent-to-Agent）通信规范73% similarUnverifiedGoogle DeepMind于2023年发布的RT-2首次验证了在550亿参数的视觉语言模型上微调机器人控制的可行性，其泛化能力相比前代提升近一倍72% similarUnverifiedGoogle DeepMind的Mariner项目也在探索类似Computer Use的方向71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/879174API
curl https://kongchang.com/api/v1/knowledge/claims/879174MCP
get_claim(id=879174)