arXiv cs.AI· Kaustuv Mukherji, Jaikrishna Manojkumar Patil, Colton Payne, Paulo Shakarian, Dana Warmsley, Nigel Stepp, Evelyn Kim·· 14 小时前AI 评分34
EntailLLM：用逻辑编程与领域知识验证智能体生成的漏洞发现路径
EntailLLM: Verifying Agent-Generated Vulnerability Discovery Paths with Domain Knowledge via Logic Programming
AI 导读
EntailLLM 框架通过独立知识图谱与带注释逻辑的逻辑程序，对 LLM 智能体提出的函数调用图遍历路径做蕴含校验，在 3 个 CWE 类别、8 个规模 405 至 12,696 个函数的二进制上，将汇总蕴含率从 77% 提升至 98%，118 种配置中仅 3 种下降。表现最佳的智能体结合了自我精炼与先验领域知识。该框架已在真实医疗设备二进制上端到端部署，无需逐设备调优即达 98% 汇总蕴含率。
正文 · 原文
Abstract:Large language model agents are increasingly used to analyze potential software vulnerabilities, but their outputs can silently violate domain knowledge, limiting their reliability in safety-critical settings such as medical devices. Prior work either treats that output as a prediction to be scored or constrains it to walks within a single knowledge graph; neither checks whether reasoning over a binary is entailed by an independent body of domain knowledge. We present EntailLLM, a framework that includes a specialized LLM agent and a validation mechanism. The agent proposes a traversal path of the function call graph for a vulnerability, which is checked against a separate knowledge graph and a logic program by an entailment procedure under annotated logic. We evaluate EntailLLM by examining six specialized agents across three CWE classes, and eight binaries varying in size from 405 to 12,696 functions. The agents improve pooled entailment from 77% to 98%, with entailment decreasing in only 3 of 118 configurations. We found that the best-performing agent combines self-refinement and a priori domain knowledge. EntailLLM is deployed end-to-end on real medical-device binaries, reaching 98% pooled entailment without per-device tuning. Our framework inherits the formal guarantees of generalized annotated logic, providing logical verification of agent output that is both explainable and grounded in well-defined semantics.
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO); Software Engineering (cs.SE) |
| Cite as: | arXiv:2608.01763 [cs.CR] |
| (or arXiv:2608.01763v2 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2608.01763 arXiv-issued DOI via DataCite |
Submission history
From: Kaustuv Mukherji [view email]
[v1]
Mon, 3 Aug 2026 06:34:31 UTC (2,262 KB)
[v2]
Sun, 25 Oct 2026 16:21:06 UTC (2,468 KB)
来源：arXiv cs.AI · arxiv.org