Metraipolis › archive › dc_70f08fded4e0

From Weak Task Specifications to Scientific Extraction Agents: Optimizing Task Construction — arXiv:2609.34829

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.34829
License (as archived)
CC0-1.0
Archived
2026-10-07T21:53:32.932Z
SHA-256 of the archived text
7bb7b0f305bae2f5704b90f3ea089b9c9f13b1397404c2be05d4a9f5211a66c9
Stable URL of this page
https://metraipolis.com/documento/dc_70f08fded4e0

How to cite

From Weak Task Specifications to Scientific Extraction Agents: Optimizing Task Construction — arXiv:2609.34829 — arXiv (en), https://arxiv.org/abs/2609.34829 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_70f08fded4e0 (Metraipolis archive dc_70f08fded4e0, sha256 7bb7b0f305bae2f5704b90f3ea089b9c9f13b1397404c2be05d4a9f5211a66c9)
BibTeX
@misc{dc70f08fded4e0,
  title = {From Weak Task Specifications to Scientific Extraction Agents: Optimizing Task Construction — arXiv:2609.34829},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.34829},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_70f08fded4e0; sha256 7bb7b0f305bae2f5704b90f3ea089b9c9f13b1397404c2be05d4a9f5211a66c9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34829. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Most methods that optimize LLM prompts and agent workflows assume that task-specific output schemas, extraction instructions, and evaluation criteria are predefined. For scientific extraction agents, however, a short task goal may not fully determine these components, while specifying them manually is costly. We study the upstream problem of constructing the task-specific configuration from a weak specification containing only a short goal and unannotated reference documents. Rather than treating automatic construction as a fixed preprocessing step, our framework constructs a task-specific schema, extraction instructions, and base training rubrics, then keeps schema construction and extraction instructions editable during optimization. Failure-focused updates concentrate textual-gradient feedback on lower-scoring documents, while training-time evaluation criteria adapt to recurring failures. On a heterogeneous-catalysis literature corpus, automatic construction remains improvable, and optimizing both schema construction and extraction instructions performs best across all four judge-rubric settings, with ablations and blinded human evaluation supporting the proposed formulation.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.34829 — 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 2026-10-07T21:27:08.642Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).