跳到正文
arXiv:cs.AI· Gaston Besanson·· 3 小时前

概率感知与确定性权威:将模型产生的观测纳入充分性校验的治理合约

Probabilistic Sensing, Deterministic Authority: Admitting Model-Produced Observations into Sufficiency-Checked Governance Contracts

AI 导读

SARC 系列第六篇论文提出将模型输出仅作为带分数的观测记录纳入治理合约:准入策略在声明假阳性上限下把分数映射为 true/false/unknown,unknown 即拒绝,感知改变裁决的概率由合约被感知字段的误纳率与 unknown 率之和上界约束。

正文

View PDF HTML (experimental)

Abstract:When a field that an authority contract needs exists only in unstructured evidence, a model can sense it. We admit the model's output only as an observation record with a score. An admission policy, with thresholds fitted on a held-out split at a declared false-positive ceiling, maps each score to true, false or unknown. Unknown denies. A deterministic, sufficiency-checked contract decides. The probability that sensing changes the verdict is bounded by the sum, over the contract's sensed fields, of the admitted-wrong and unknown rates. This is an instantiation of union-bound reasoning, indexed by the contract. Minimising the estimated bound is a valid cost model for choosing among sufficient contracts. In a registered study on two constructed domains with two sensor families (36,000 model calls), no cell refuted the bound. Deny-to-allow changes from sensing appeared for the first time in this programme: 13 of 21,000 test verdicts, all from 3 contradictory records; each flip in a cell with a registered bound lay under it. Sensing-aware selection picked the lower-exposure contract in 4 of 4 registered tests. Both sensors' scores were informative but not calibrated. Correctness is relative to the declared loss model, candidate representation and reachable states; all domains are constructed.
Comments: 23 pages. Sixth paper of the SARC series. Preregistered; external reviews and an independent reproduction are committed in the repository. Code, caches and checkers: this https URL (tag v1.0.1), archived at DOI https://doi.org/10.5281/zenodo.23224016
Subjects: Software Engineering (cs.SE); Artificial Intelligence (cs.AI)
Cite as: arXiv:2610.10978 [cs.SE]
  (or arXiv:2610.10978v1 [cs.SE] for this version)
  https://doi.org/10.48550/arXiv.2610.10978

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Carlos Gaston Besanson [view email]
[v1] Wed, 7 Oct 2026 23:02:20 UTC (55 KB)

来源:arXiv:cs.AI · arxiv.org