AgenticGenTAMP: coding agents synthesize reusable robot TAMP policies across 28 envs
tomssilver · x · 2026-09-25
Tom Silver's group at Princeton and collaborators present AgenticGenTAMP, using coding agents to synthesize reusable programmatic policies for generalized task and motion planning, evaluated across 28 simulated environments.
Key points:
- Generalized TAMP exploits regularities across instances (varying object counts, configurations, geometry) so one program covers them all, via learned samplers, feasibility predictors, search heuristics, or abstractions
- Strict setup: the agent gets only task descriptions and state/action spaces, a $20 model-usage budget per synthesis run, no network or host filesystem, isolated Docker execution, and no environment source code—just a client API
- Paper, code, and videos are public
More from Embodied
- Perceptron releases Mk1.5, one model to fly drones and control robots — AkshatS07 · 2026-09-25
- Robots doing unexpectedly clever things: problems unsolvable six months ago now cracked — tomssilver · 2026-09-25
- ROS 2 Makoa Charts a Less-is-More Path: Cutting, Not Adding — facontidavide · 2026-09-25
- Sims and Sensors Called the Most Fertile Ground at the Infra Layer — pzakin · 2026-09-25
- Robot simulation is the cleanest anti-gaming incentive, argues Bittensor SN49 — bittingthembits · 2026-09-25
- Princeton-led work uses coding agents to synthesize reusable policies for generalized task and motion planning — tomssilver · 2026-09-25