Megadose AI progress, ranked and analyzed.

Coding Agents for Generalized Task and Motion Planning Problems

· HF Daily Papers ·
Coding agents beat available hand-engineered TAMP planners on mean success while using far less per-instance compute.

The paper tests Claude Code and Codex variants on 28 simulated robotics environments from KinDER and PDDLStream. The agents synthesize programs under a fixed budget, then those programs are frozen and run on unseen instances. Across 98,000 evaluation episodes, the agent-written programs reached 56% to 95% mean success on environments with planner baselines, versus 47% for the planners. The authors say the agents’ logs show iterative testing: calibrating physical models, checking edge cases, and revising strategies. HF Daily Papers' note

score 5

Categories: Research