Metraipolis › archive › dc_e6e8df082ea8

Learning Clinical-Trial Strategy: Offline Policy Training for Decision Agents — arXiv:2608.03606

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.03606
License (as archived)
CC0-1.0
Archived
2026-10-31T12:16:29.441Z
SHA-256 of the archived text
319e4a76ebb2dab037f2dc9d4a9deac0c107a7368c2ce7f310d75e1471c82c7a
Stable URL of this page
https://metraipolis.com/documento/dc_e6e8df082ea8

How to cite

Learning Clinical-Trial Strategy: Offline Policy Training for Decision Agents — arXiv:2608.03606 — arXiv (en), https://arxiv.org/abs/2608.03606 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e6e8df082ea8 (Metraipolis archive dc_e6e8df082ea8, sha256 319e4a76ebb2dab037f2dc9d4a9deac0c107a7368c2ce7f310d75e1471c82c7a)
BibTeX
@misc{dce6e8df082ea8,
  title = {Learning Clinical-Trial Strategy: Offline Policy Training for Decision Agents — arXiv:2608.03606},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.03606},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e6e8df082ea8; sha256 319e4a76ebb2dab037f2dc9d4a9deac0c107a7368c2ce7f310d75e1471c82c7a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.03606. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Clinical development is sequential decision-making under uncertainty, where a sponsor must plan a portfolio of experiments from heterogeneous evidence. We study this setting by framing oncology clinical development as an offline decision-making problem in which an agent predicts the next six-month trial portfolio of an oncology drug program from information available at the decision date. To support this, we construct a temporal dataset that combines 31.7k heterogeneous public data records, including trial registries, regulatory reviews, sponsor filings, utilization data, and epidemiology, into 881 offline decision episodes across 45 historical programs. We compare four offline objectives: behavioral cloning, reward-weighted behavioral cloning, learned-reward training, and value-based implicit Q-learning against four frontier LLM agents that share a common date-gated retrieval scaffold across held-out drug, sponsor, drug-class, and temporal splits. Models trained offline outperform the non-fine-tuned baselines, particularly in the post-August 2025 contamination-clean holdout. Reward-weighted behavioral cloning performs the best, obtaining 46.2% indication F1 and 14.2% strict F1 against 25.0% and 2.1%, respectively, for the best-performing tool agent on each metric. These results suggest that structured offline learning can teach agents to plan clinical experiments.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.03606 — 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 2026-10-31T12:05:52.821Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).