ENPIRE harness automates robotics research, hitting 99% success on dexterous tasks
chris_j_paxton · x · 2026-10-07
RoboPapers Episode 110 covers ENPIRE, a harness built by @wenlixiao, @jiaxiejason and @TongheZhang01 that captures physical feedback and enables automated robotics research driven by frontier LLMs. It minimizes human effort while training policies that achieve a 99% success rate on challenging dexterous tasks like organizing a pin box and fastening zip ties. Panelists including Chris Paxton discuss what role remains for human researchers when 'autoresearch in the physical world' becomes plausible.
More from Embodied
- Robotics company LeRobot jokes it now makes music videos too — Thom_Wolf · 2026-10-08
- MIT's SOLE-R1: Video-Language Reasoning as the Sole Reward Enables Zero-Shot On-Robot RL — micoolcho · 2026-10-08
- 200 builders revive vintage toys as voice agents at AssemblyAI hackathon — AssemblyAI · 2026-10-07
- Robotics is having its moment: why roboticists are Renaissance generalists — lukas_m_ziegler · 2026-10-07
- PhD student wraps up robotics planning series with EMPIRIC residual world model — tomssilver · 2026-10-07
- Wearing a BCI headband like Chinese quant traders is the new productivity hack — alexbilz · 2026-10-07