arXiv cs.MA· Kardelen Subasi, Jaros{\l}aw A. Chudziak·· 13 小时前AI 评分31
BiasGuard-D：用类型化对抗契约实现可审计的 LLM 输出质量管控
Conflicted Records: Typed Adversarial Contracts for Auditable LLM Output Quality Enforcement
AI 导读
BiasGuard-D 通过两个来自不同模型族的对抗角色交换类型化声明文档、三模型法定人数多数投票裁决，并用指称加语义的组合不变量校验结果，从而捕捉单纯指称检查漏掉的缺陷。在 HaluEval QA 上，该审议层仅调整角色设定即可从刻板印象修复迁移到幻觉修复，正确答案恢复率提升 49 个百分点。每条声明、投票和指令都以类型化记录存储，为受新兴 AI 监管约束的运营方提供可审计的溯源链。
正文 · 原文
Abstract:LLM-generated records increasingly populate production information systems where they are treated as data assets subject to the same quality expectations as records from operational databases. A single such record can simultaneously satisfy and violate a quality rule: a grounded fact adjacent to a hallucination, a useful answer adjacent to a stereotype. Classical data cleaning cannot resolve this because the two judgments are incompatible over the same span; a single LLM judge cannot either, since it shares the training biases that produced the conflict. Existing interventions such as prompt engineering, chain-of-thought, and RLHF reduce surface error rates but provide no audit trail and cannot distinguish genuine remediation from blanket suppression. The question is whether a deterministic check over structured inter-agent records can catch the defects that referential checks alone miss. We present BiasGuard-D: two adversarial roles drawn from distinct model families exchange typed claim documents, a three-model quorum adjudicates by majority vote, and a combined referential-plus-semantic invariant validates the verdict. The semantic component fires on verdicts referential checks miss; removing the iterative loop drops invariant satisfaction sharply. The deliberation layer transfers from stereotype to hallucination remediation on HaluEval QA with only role-framing changes, raising right-answer recovery by 49 percentage points; the triage gate required a cross-family second auditor to maintain recall on objective hallucination. Every claim, vote, and instruction is stored as a typed record giving operators a provenance trail auditable under emerging AI regulations.
| Comments: | Accepted at the International Conference on Cooperative Information Systems (CoopIS 2026) |
| Subjects: | Databases (cs.DB); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2610.29669 [cs.DB] |
| (or arXiv:2610.29669v1 [cs.DB] for this version) | |
| https://doi.org/10.48550/arXiv.2610.29669 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Kardelen Subaşı [view email]
[v1]
Wed, 21 Oct 2026 12:43:35 UTC (84 KB)
来源：arXiv cs.MA · arxiv.org