arXiv cs.AI· Hamed Khosravi, Xiaoming Huo·· 12 小时前AI 评分38
PRIME：让 LLM 智能体基于可验证证据进行推理
PRIME: Towards Verifiable Evidence-Grounded Reasoning in LLM Agents
AI 导读
PRIME（Precommitted Rules for Inference from Minimal Evidence）通过预先固定证据路径、由独立程序记录每次检查结果并拒绝无依据结论，让 LLM 智能体只报告规则与账本授权的结论。
正文 · 原文
Abstract:LLM agents are used to check records against an organization's requirements, for example whether a vulnerability report is ready for investigation. Requirements often allow several ways to be met, such as a severity rating with a curator-reviewed advisory or a published advisory link. Agents fail by accepting a report after a required check has failed, or by settling for a weaker conclusion without trying another way that could still succeed. We introduce PRIME (Precommitted Rules for Inference from Minimal Evidence). Before checking, it fixes evidence routes specifying how each conclusion may be justified. A separate program records each check result and refuses any conclusion the results do not justify. It also points the agent to checks that could justify a stronger conclusion, or runs them itself. PRIME reports only conclusions authorized by its rules and ledger. With correct requirements, exact checks at known locations, and sufficient budget at stopping, completion reaches the strongest supported conclusion. On 480 public-record tasks, an agent shown the requirements reported unjustified conclusions on 26.7% of tasks, while PRIME reported none. Its directions and added checks raised the average score for justified conclusions from 30.1, with refusal alone, to the maximum of 59.2 out of 100.
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.22426 [cs.AI] |
| (or arXiv:2610.22426v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.22426 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Hamed Khosravi [view email]
[v1]
Thu, 15 Oct 2026 05:03:22 UTC (226 KB)
来源：arXiv cs.AI · arxiv.org