Brohan and colleagues: RT-2 vision-language-action robotic control

What changed

Co-fine-tuned vision-language models on web data and robot trajectories while representing robot actions as tokens, reporting transfer to new robotic tasks.

Scientific scope

Co-fine-tuned vision-language models on web data and robot trajectories while representing robot actions as tokens, reporting transfer to new robotic tasks. Scope limitation: RT-2 results concern the studied robots, tasks and setups; they do not prove general-purpose real-world robotic reasoning.

Evidence

The retained primary-source abstract, bibliographic record and stated review scope support this provisional assessment. No independent experimental replication by WRS is claimed. RT-2 results concern the studied robots, tasks and setups; they do not prove general-purpose real-world robotic reasoning.

Contributors

• The named authors listed in the linked research record · SCIENCE: Complete source byline retained in the research record; human research direction, method development and evaluation. Individual labor shares are not measured.

• Brohan and colleagues: RT-2 vision-language-action robotic control — original paper
• Brohan and colleagues: RT-2 vision-language-action robotic control — official research page

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction