Metraipolis › archive › dc_20e2ebed228d

Risk-Controlled Selective LLM Answering by Pricing Label-Free Checks — arXiv:2609.37493

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.37493
License (as archived)
CC0-1.0
Archived
2026-10-07T16:27:19.128Z
SHA-256 of the archived text
e78671c1aa5c14b5d4254089cc138d1daeb323bc04662129a5b686beb6f64e7c
Stable URL of this page
https://metraipolis.com/documento/dc_20e2ebed228d

How to cite

Risk-Controlled Selective LLM Answering by Pricing Label-Free Checks — arXiv:2609.37493 — arXiv (en), https://arxiv.org/abs/2609.37493 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_20e2ebed228d (Metraipolis archive dc_20e2ebed228d, sha256 e78671c1aa5c14b5d4254089cc138d1daeb323bc04662129a5b686beb6f64e7c)
BibTeX
@misc{dc20e2ebed228d,
  title = {Risk-Controlled Selective LLM Answering by Pricing Label-Free Checks — arXiv:2609.37493},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.37493},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_20e2ebed228d; sha256 e78671c1aa5c14b5d4254089cc138d1daeb323bc04662129a5b686beb6f64e7c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37493. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Serving an answer from a large language model requires deciding when to abstain, yet a verifier's ranking accuracy alone does not determine the error rate among served answers. We introduce PriceCheck, which builds a compact family of decision rules from label-free checks such as re-solving a problem. Each check has a price: its agreement rates on correct and incorrect answers and its cost per run. Prices fitted on a small, class-enriched labelled set compose into predictions of a schedule's coverage and cost, guiding which checks to run and when to stop. A calibration test then selects a schedule at a stated selective-risk target. In mathematics, the selected schedules serve 76.1% of answers on average and keep held-out selective risk below 1.5% on all 15 splits. Under the shared testing protocol, PriceCheck serves more answers at that target than reward models, a prompted judge, the generator's confidence and a trained correctness classifier. At matched coverage, it keeps the fewest wrong answers among these scorers. Across 118 diagnostic schedules, price-based coverage predictions have a rank correlation of 0.97 with observed coverage. These results show that choosing how checks are combined and stopped matters alongside how well a verifier ranks answers. Code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.37493 — 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 2026-10-07T15:54:53.986Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).