Metraipolis › archive › dc_026ed4fd711a

Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI — arXiv:2609.38143

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.38143
License (as archived): CC0-1.0
Archived: 2026-10-07T16:24:31.998Z
SHA-256 of the archived text: 9c1eb4948fdce9cd56fe963314c5c4c64d52503fed1d5a2624918ba3fe00e2fa
Stable URL of this page: https://metraipolis.com/documento/dc_026ed4fd711a

How to cite

Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI — arXiv:2609.38143 — arXiv (en), https://arxiv.org/abs/2609.38143 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_026ed4fd711a (Metraipolis archive dc_026ed4fd711a, sha256 9c1eb4948fdce9cd56fe963314c5c4c64d52503fed1d5a2624918ba3fe00e2fa)

BibTeX

@misc{dc026ed4fd711a,
  title = {Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI — arXiv:2609.38143},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.38143},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_026ed4fd711a; sha256 9c1eb4948fdce9cd56fe963314c5c4c64d52503fed1d5a2624918ba3fe00e2fa},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38143. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Agent performance depends on both reasoning ability and the environment in which it acts. We study test-time AI-for-AI, asking how a Builder can learn to construct better execution environments for a Target while both models' weights remain fixed. To make the Builder's experience reusable, we introduce Meta-Skill: principles specifying when support is needed and what resources to provide. The Builder learns these principles from Target's execution feedback on the development set, then uses the frozen skill bank to construct harnesses for unseen tasks. Across Harness-Bench and NewtonBench, full-bank meta-skills improve macro-average performance by 8.95 percentage points over no-skill construction, and 12.02 points over direct delivery of the same bank to the Target. These results highlight the value of translating experience into executable support. Gains when the same model serves both roles further suggest a path to system level self-improvement through learning to build better environments.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.38143 — 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 2026-10-07T15:54:53.694Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).