Metraipolis › archive › dc_4196cf2afbfb

Learning from Research: Toward Lifelong Agent Harness Evolution — arXiv:2609.40169

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.40169
License (as archived): CC0-1.0
Archived: 2026-10-07T12:43:08.105Z
SHA-256 of the archived text: 92cf31f37eb90e45ec7e8f4689c11f2d0dfb72b49ea2d989b6091ac7c0b47af8
Stable URL of this page: https://metraipolis.com/documento/dc_4196cf2afbfb

How to cite

Learning from Research: Toward Lifelong Agent Harness Evolution — arXiv:2609.40169 — arXiv (en), https://arxiv.org/abs/2609.40169 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4196cf2afbfb (Metraipolis archive dc_4196cf2afbfb, sha256 92cf31f37eb90e45ec7e8f4689c11f2d0dfb72b49ea2d989b6091ac7c0b47af8)

BibTeX

@misc{dc4196cf2afbfb,
  title = {Learning from Research: Toward Lifelong Agent Harness Evolution — arXiv:2609.40169},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.40169},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4196cf2afbfb; sha256 92cf31f37eb90e45ec7e8f4689c11f2d0dfb72b49ea2d989b6091ac7c0b47af8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.40169. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Language agents are expected to solve increasingly complex tasks, creating a growing need for continual improvement. One promising approach is to evolve the agent harness, the software that governs tool use, memory management, and task execution, while keeping the underlying language model fixed. Recent methods automate this process by using a meta coding agent to modify the harness based on execution feedback. However, relying on that agent's existing knowledge and observed failures can restrict exploration and make adaptation reactive. Inspired by how human experts learn from the research literature for new solutions, we introduce ScholarEvolve, a framework that automatically draws on state-of-the-art research to guide harness evolution. ScholarEvolve organizes the harness evolution directions into functional modules and uses topic modeling to identify distinct improvement strategies for each module. It implements these strategies and evaluates their combinations to improve task performance. Moreover, the framework is designed to incorporate new publications over time, allowing research advances to drive proactive lifelong evolution. Experiments demonstrate improvements on AppWorld and Tau2-Bench. ScholarEvolve raises Qwen3.5-27B task goal completion from 49.6% to 63.6% on AppWorld Challenge, and raises GPT-5.4-mini pass@1 from 72.7% to 81.9% on Tau2-Bench Telecom.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.40169 — 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 2026-10-07T12:14:13.814Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).