Metraipolis › archive › dc_deb2c5808362

ScienceCUA: Resource-to-Skill Construction and Execution-Guided Skill Evolution for Scientific Computer Use Agents — arXiv:2611.03238

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.03238
License (as archived)
CC0-1.0
Archived
2026-11-03T18:14:27.481Z
SHA-256 of the archived text
0c44c8ae2257e0c96703ed642606035d2e62905879b394b76a8037cc669ee704
Stable URL of this page
https://metraipolis.com/documento/dc_deb2c5808362

How to cite

ScienceCUA: Resource-to-Skill Construction and Execution-Guided Skill Evolution for Scientific Computer Use Agents — arXiv:2611.03238 — arXiv (en), https://arxiv.org/abs/2611.03238 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_deb2c5808362 (Metraipolis archive dc_deb2c5808362, sha256 0c44c8ae2257e0c96703ed642606035d2e62905879b394b76a8037cc669ee704)
BibTeX
@misc{dcdeb2c5808362,
  title = {ScienceCUA: Resource-to-Skill Construction and Execution-Guided Skill Evolution for Scientific Computer Use Agents — arXiv:2611.03238},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03238},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_deb2c5808362; sha256 0c44c8ae2257e0c96703ed642606035d2e62905879b394b76a8037cc669ee704},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03238. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Computer-use agents for scientific software require application-specific expertise. Much of this expertise is documented in human-authored software manuals and web pages. However, turning these resources into executable, reusable skills remains challenging due to mismatches between documented and deployed software versions and differences in scope between documentation coverage and task requirements. We introduce ScienceCUA, a two-stage framework combining resource-to-skill construction with execution-guided skill evolution. First, ScienceCUA identifies tasks supported by the software from its documentation, infers operational trajectories, and generalizes them into reusable Skills. Command catalogs, syntax constraints, and shortcuts are organized separately as References for on-demand lookup. Then, ScienceCUA uses execution feedback to drive skill self-evolution by contrasting successful and failed trajectories for the same task, identifying deficiencies, revising existing skills or adding new ones, and updating the skill library only after repeated validation in real environments. To enable reliable evaluation without benchmark noise or task leakage, we introduce two improved benchmarks. ScienceBoard-Verified provides an audited and corrected basis for measuring task performance, while ScienceBoard-VE separates skill evolution from final testing through disjoint evolution and test tasks and further supports native-agent evaluation. Experiments show that the initial Skills and References improve average task success rate by up to 7.23 percentage points over the base model, while skill self-evolution increases the total gain over the baseline to as much as 17.12 percentage points. ScienceCUA turns software documentation into an evolving set of executable skills and queryable references, offering a practical way for general-purpose agents to use scientific software reliably and at scale.

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/2611.03238 — 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 2026-11-03T18:09:53.450Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).