Metraipolis › archive › dc_fb60ace5d236

ScienceClaw: Benchmarking Continual Self-Evolution of AI-for-Science Agents Across the Natural and Social Sciences — arXiv:2610.08691

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.08691
License (as archived)
CC0-1.0
Archived
2026-10-07T11:19:51.877Z
SHA-256 of the archived text
878a2b086768fb7e3a88c120ca450beec8a2b0f8a7a43c7d206c717f3009150e
Stable URL of this page
https://metraipolis.com/documento/dc_fb60ace5d236

How to cite

ScienceClaw: Benchmarking Continual Self-Evolution of AI-for-Science Agents Across the Natural and Social Sciences — arXiv:2610.08691 — arXiv (en), https://arxiv.org/abs/2610.08691 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fb60ace5d236 (Metraipolis archive dc_fb60ace5d236, sha256 878a2b086768fb7e3a88c120ca450beec8a2b0f8a7a43c7d206c717f3009150e)
BibTeX
@misc{dcfb60ace5d236,
  title = {ScienceClaw: Benchmarking Continual Self-Evolution of AI-for-Science Agents Across the Natural and Social Sciences — arXiv:2610.08691},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.08691},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fb60ace5d236; sha256 878a2b086768fb7e3a88c120ca450beec8a2b0f8a7a43c7d206c717f3009150e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.08691. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language model agents are accelerating scientific automation, yet verified executions rarely become persistent program-level improvements, and existing evaluations do not examine this process across sequential tasks in both the natural and social sciences. We formalize ScienceClaw as fixed-parameter program self-evolution that unifies task solving, scientific verification, and program updates. ScienceClaw-Eval spans 23 disciplines and measures scientific correctness, evolutionary gain, retention, cross-dataset transfer, and evolution cost through sequential streams and independent reset evaluation. Our framework repairs executable workflows through multi-turn interaction, converts re-execution-verified failure--success trajectories into linked Skill and Operator candidates, and retains an update only when source-task replay reproduces the repair and independent scientific tasks improve. Code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.08691 — 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 2026-10-07T11:15:24.628Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).