robotics
The Robot Calls the Planner Only When the Routine Breaks
ChunkAct compiles remembered robot routines into action chunks, reserving language-model calls for decisions and failures.
Summary
ChunkAct compiles remembered robot routines into action chunks, reserving language-model calls for decisions and failures.
Agentic robot systems can ask a language model to reason before every primitive action, repeatedly rereading the same history. ChunkAct's authors instead compile memory into chains of primitives that execute autonomously until a failure returns control to the planner. On LIBERO-Pro, they report a five-point success-rate gain over their agentic-robotics baseline while using 2.9 times fewer model calls, 4.4 times fewer input tokens and 2.3 times fewer output tokens. These are benchmark results, but the design isolates a practical systems idea: use expensive reasoning at decision boundaries, not at every movement.
Why it matters
ChunkAct compiles remembered robot routines into action chunks, reserving language-model calls for decisions and failures.
Limits and context
- These are benchmark results, but the design isolates a practical systems idea: use expensive reasoning at decision boundaries, not at every movement.
Key claims
ChunkAct compiles remembered robot routines into action chunks, reserving language-model calls for decisions and failures.
Qualification: These are benchmark results, but the design isolates a practical systems idea: use expensive reasoning at decision boundaries, not at every movement.
Evidence: source-2026-10-14-003
Sources
- arXiv preprint 2610.18874, original v1arXiv / submitting research authors · primary research
Corrections
No corrections have been recorded for this story.