Metraipolis › archive › dc_aaae122d0971

From Tool State to Scientific Action: Auditing Decision Provenance in Biomedical Research Agents — arXiv:2611.01527

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.01527
License (as archived)
CC0-1.0
Archived
2026-11-03T10:22:15.623Z
SHA-256 of the archived text
39fb3f8735805fe0e023b6327d9036e050dd8ede15c9a8cd1dc38dcf670fd7d2
Stable URL of this page
https://metraipolis.com/documento/dc_aaae122d0971

How to cite

From Tool State to Scientific Action: Auditing Decision Provenance in Biomedical Research Agents — arXiv:2611.01527 — arXiv (en), https://arxiv.org/abs/2611.01527 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_aaae122d0971 (Metraipolis archive dc_aaae122d0971, sha256 39fb3f8735805fe0e023b6327d9036e050dd8ede15c9a8cd1dc38dcf670fd7d2)
BibTeX
@misc{dcaaae122d0971,
  title = {From Tool State to Scientific Action: Auditing Decision Provenance in Biomedical Research Agents — arXiv:2611.01527},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.01527},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_aaae122d0971; sha256 39fb3f8735805fe0e023b6327d9036e050dd8ede15c9a8cd1dc38dcf670fd7d2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.01527. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A scientific agent can give the same verdict while reading and citing different evidence. BioAgent-DF tests this separation using paired reference and candidate values saved at the same scientific computation site. It restores or transplants those values at their original coordinates, then runs the agent on healthy hardware. Across eight SciFact batches, Qwen3.5-4B shows repeatable citation changes in three capture-claim combinations, and Gemma 3 4B IT shows repeatable read and citation changes in four. These seven combinations cover two claims. Full restoration recovers the reference evidence sets while verdicts remain unchanged. Verdict changes observed under the baseline prompt did not recur under the explicit-example prompt. Complementary tool and optimizer experiments distinguish recovery of numerical state, recovery of ranked lists, and restoration of an agent's visible history. These results show that an audit should track both the evidence an agent uses and its final answer. In hERG cardiac-safety assay allocation, silent numerical faults changed testing decisions while the fixed-budget shortlist kept the same number of compounds labeled as blockers. After the agent had used faulty values, restoring its earlier execution state and supplying certified scores recovered all 39 changed decisions, whereas replacing the prediction file recovered none; no incorrect action was committed. A streaming implementation of the arithmetic checks reduced the total time for forward execution and verification by 42.4% relative to the original FP64 implementation, with identical flags and recovery outputs.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG); Quantitative Methods (q-bio.QM).
Pagina del lavoro: https://arxiv.org/abs/2611.01527 — 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 2026-11-03T10:19:49.821Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).