Metraipolis › archive › dc_ebf312c10bc2
Decision-Centric Research: A Calibrated Decision Model and Benchmark for Scientific Agents — arXiv:2611.00774
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.00774
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T10:22:00.091Z
- SHA-256 of the archived text
83aa956c09e985cab3adbcae1505394345731352a5ae19586c9c215eb3284734- Stable URL of this page
https://metraipolis.com/documento/dc_ebf312c10bc2
How to cite
Decision-Centric Research: A Calibrated Decision Model and Benchmark for Scientific Agents — arXiv:2611.00774 — arXiv (en), https://arxiv.org/abs/2611.00774 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ebf312c10bc2 (Metraipolis archive dc_ebf312c10bc2, sha256 83aa956c09e985cab3adbcae1505394345731352a5ae19586c9c215eb3284734)
BibTeX
@misc{dcebf312c10bc2,
title = {Decision-Centric Research: A Calibrated Decision Model and Benchmark for Scientific Agents — arXiv:2611.00774},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.00774},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ebf312c10bc2; sha256 83aa956c09e985cab3adbcae1505394345731352a5ae19586c9c215eb3284734},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00774. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Scientific agents make consequential choices about evidence, next actions and research strategy, but choices embedded in generated text are difficult to evaluate independently. Decision-Centric Research (DCR) is introduced to make individual choices the common unit of supervision, prediction and evaluation through typed predictions over an explicit Research State. DCR defines 24 decision types, ten of which are instantiated in the Research Decision Benchmark (RD-Bench) of 2,000 rule-generated cases. The Decision-Centric Research Model (DCRM) shares a candidate-token readout across types to produce answers and option probabilities in one forward pass. Under a held-out research-domain split, a 0.6B DCRM (state-aware) achieves 89.4% Top-1 accuracy, compared with 27.8% for DCRM (question-only) trained with the same recipe and 36.4% for DeepSeek-V4.1-Flash. All 30 type-seed comparisons against DCRM (question-only) are significant after Holm correction. After held-out calibration, the resulting prediction sets contain a mean of 1.515 candidates at 97.1% empirical label coverage, reducing mean set size by 57.4% relative to DCRM (question-only) at similar coverage. These results establish a compact state-aware decision layer with accurate typed prediction and an inspectable uncertainty interface. 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/2611.00774 — 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 2026-11-03T10:19:49.787Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).