Agents & Tool Use · 2022
Do As I Can, Not As I Say: Grounding Language in Robotic Affordances (SayCan)
Grounded a language model’s plans in what a robot can actually do by combining the model’s task knowledge with learned affordance values, letting an LLM direct real-world robotic behavior.
Editorial record
Plain-language summary
SayCan scores candidate skills by multiplying an LLM’s estimate that a skill helps achieve the instruction with a value function’s estimate that the skill is currently feasible, so the robot picks useful and possible actions. This let a robot carry out long natural-language instructions in a kitchen. It was an influential early bridge from language models to embodied action.
Knowledge graph
Relationships
Descendants
Applies toEvidence: Direct
Language Models are Few-Shot Learners (GPT-3)
SayCan grounds a language model in a robot value function of affordances
P-521
Source record
Provenance
- Record ID
- P-521
- Record created
- 2026-07-13
- Last reviewed
- 2026-07-14
- Record version
- 2
- https://arxiv.org/abs/2204.01691
- arXiv:2204.01691
Citation caveat: Citation metadata is approximate and marked unverified in the source dataset.