Metraipolis › archive › dc_fa2b9e2d3061
TypedBench: A Benchmark for Calibration, Framing Sensitivity, and Cost in System One Decision Models — arXiv:2610.11392
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.11392
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T05:47:42.912Z
- SHA-256 of the archived text
68c3d146a7249e7641850fbfcf0440906f336f1f4f32f4689ebf8d2c3c82ab99- Stable URL of this page
https://metraipolis.com/documento/dc_fa2b9e2d3061
How to cite
TypedBench: A Benchmark for Calibration, Framing Sensitivity, and Cost in System One Decision Models — arXiv:2610.11392 — arXiv (en), https://arxiv.org/abs/2610.11392 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fa2b9e2d3061 (Metraipolis archive dc_fa2b9e2d3061, sha256 68c3d146a7249e7641850fbfcf0440906f336f1f4f32f4689ebf8d2c3c82ab99)
BibTeX
@misc{dcfa2b9e2d3061,
title = {TypedBench: A Benchmark for Calibration, Framing Sensitivity, and Cost in System One Decision Models — arXiv:2610.11392},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.11392},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fa2b9e2d3061; sha256 68c3d146a7249e7641850fbfcf0440906f336f1f4f32f4689ebf8d2c3c82ab99},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.11392. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
System One models output calibrated probabilities over typed answers such as categorical choices, ordinal levels, or binary outcomes, via a non-generative interface. Software can act on these probabilities through thresholds, cost-weighted choices, and escalation rules. Consequently, if these probabilities are miscalibrated or wording-sensitive, the software ma take unintended actions leaving human operators with no textual rationale to inspect. Current evaluations largely report accuracy and calibration on public classification datasets without a clear reference. We present TypedBench, a benchmark for typed decision models like Jev built from seven policy-labelled generators and nine evaluation suites. We report accuracy as median and range across paraphrases, and calibration error relative to the finite-sample noise floor of a matched, perfectly calibrated predictor. We assess probability quality through selective prediction, ordinal proper scoring rules, and realised cost under asymmetric cost matrices. We evaluate a hosted model, an open encoder, and a family of open decoders spanning 0.8B-9B parameters on identical items. The hosted model follows the stated policy but is wording-sensitive and systematically underconfident; under asymmetric costs, using its probabilities can be worse than taking its top answer. Decoders route exactly and are slow as options or questions are added. The decoder is least accurate on policy questions and degrades with more options. Overall, typed decision models must be evaluated jointly on policy adherence, wording robustness, probability quality, and induced decision outcomes. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Applications (stat.AP). Pagina del lavoro: https://arxiv.org/abs/2610.11392 — 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 2026-10-09T05:47:00.705Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).