KAIST's QCP lets robot policies control execution quality in real time
kaist-ai · hf · 2026-11-02
A KAIST/LGE/Microsoft Research paper introduces Quality-Controllable Policy (QCP), enabling inference-time control of individual quality attributes (speed, smoothness) and their combinations. An iterative hindsight relabeling loop collects online rollouts under target quality labels, relabels them by measured quality, and fine-tunes. On Robomimic and RoboEval, QCP consistently improves both quality and success rates, exceeding initial demonstration quality, with real-world validation.
More from Embodied
- Autonomous's Harness Pro uses an ESP32-P4, not a mini PC, to monitor AI coding agents — dee_hw · 2026-11-02
- ANYbotics begins customer deliveries of ANYmal X, first legged robot certified for explosive atmospheres — lukas_m_ziegler · 2026-11-02
- InSpatio opens China's first embodied simulation data center targeting 1M hours/month — 智东西 · 2026-11-02
- GR-BC mines better trajectories from imperfect demos, hitting ~91% success on precise manipulation — 机器之心 · 2026-11-02
- RAM prices tripled, so Tesla cut a third of memory from Optimus's brain — aakashgupta · 2026-11-02
- Too lazy on weekends? Watch this robot tidy up a desktop — flngr · 2026-11-02