Unverified70% confidenceFactExact time
RT-2 (Robotics Transformer 2) directly uses a vision-language model to generate robot control instructions end-to-end.
2
Sources
70%
Confidence
Long-term
Relevance
7/22/2026
First Seen
Sources
Related Claims
VerifiedRT-2 first demonstrated the feasibility of the VLA architecture by discretizing robot actions into tokens and having the vision-language model directly output action sequences.69% similarVerifiedVLA directly connects pre-trained vision-language models with robot action policies, enabling robots to process natural language instructions, camera images, and low-level motion control in a unified manner.67% similarUnverifiedThe Transformer is centered on the Self-Attention mechanism, consists of an encoder and a decoder, and is capable of processing information from all positions in a sequence in parallel.63% similarUnverifiedThe recommended Transformer learning path is: first build intuition with interactive attention visualization tools, then understand the architecture through layer-by-layer animation tutorials, and finally implement the corresponding tensor operations through code.63% similarUnverifiedFigure's robots and Tesla's Optimus are being tested on tasks such as sorting and simple assembly, using vision-language-action models to map camera images and voice commands into action sequences.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/586701API
curl https://kongchang.com/api/v1/knowledge/claims/586701MCP
get_claim(id=586701)