Metraipolis › archive › dc_bbc83ddbb7ea

HarnessPAI: An Evolving Harness for Physical AI — arXiv:2609.29166

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.29166
License (as archived): CC0-1.0
Archived: 2026-10-08T04:27:20.287Z
SHA-256 of the archived text: c2a171aa8b6b6d1873d82fc62d105f4e4dd0f4aaf4e66663c747de05e6a5efd3
Stable URL of this page: https://metraipolis.com/documento/dc_bbc83ddbb7ea

How to cite

HarnessPAI: An Evolving Harness for Physical AI — arXiv:2609.29166 — arXiv (en), https://arxiv.org/abs/2609.29166 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bbc83ddbb7ea (Metraipolis archive dc_bbc83ddbb7ea, sha256 c2a171aa8b6b6d1873d82fc62d105f4e4dd0f4aaf4e66663c747de05e6a5efd3)

BibTeX

@misc{dcbbc83ddbb7ea,
  title = {HarnessPAI: An Evolving Harness for Physical AI — arXiv:2609.29166},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.29166},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bbc83ddbb7ea; sha256 c2a171aa8b6b6d1873d82fc62d105f4e4dd0f4aaf4e66663c747de05e6a5efd3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29166. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Physical AI aims to build embodied agents that perceive the world, understand and reason about it, and decide how to act. Yet the field has focused primarily on the last component: the action model that maps observations to low-level controls. The prevailing training recipe can erode the perceptual and reasoning capabilities needed for robust behavior, leaving even strong action models vulnerable to scene perturbations and long-horizon tasks. We introduce HarnessPAI, a model- and embodiment-agnostic Harness framework for Physical AI that treats code as the executable and evolvable interface that organizes the underlying action primitive. The framework separates two timescales: within a rollout, it executes open-loop at the program level, with a fixed program guiding and checking execution; across rollouts, it evolves closed-loop, using execution feedback to revise the program and distill failures into reusable skills. Across desktop robot arms, household robots, a robot vacuum, and a legged walking agent, HarnessPAI improves on both pure action models and code-as-policy baselines without retraining the underlying model: a 61.6-point gain over $\pi_{0.5}$ on LIBERO-PRO and a 27.2-point gain over WorldDreamer on RoboCasa atomic tasks. Once a program is selected, rollout execution requires no online high-level LLM deliberation. Beyond execution, the converged program is also a cheap and reliable expert-data collector, and fine-tuning $\pi_{0.5}$ on collected expert data lifts success rate on LIBERO-PRO by 38.8 points. Our results suggest that the frontier of Physical AI depends not only on stronger action models, but also on executable harnesses that integrate perception, task understanding and reasoning, and action execution into a unified, verifiable, and feedback-driven system. Website: this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.AI).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.29166 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T03:50:58.017Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).