arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2605.09421cs.SE
收藏

MACAA:基于信念修正的多智能体推理用于代码作者身份验证

MACAA: Belief-Revision Multi-Agent Reasoning for Code Authorship Verification

Nankai University(南开大学);Runjian Co., Ltd.(润建股份有限公司)

Jingwei Ye, Zhi Wang, Xin Li, Cong Gao, Chenbin Su, Jieshuai Yang, Jianfei Tang, Ge Chu

AI总结:

MACAA提出一种无需训练的多智能体框架,通过协调器和四个专家代理分析代码证据,提升代码作者身份验证的准确性和可审计性,实现89.15%的F1分数。

摘要:

代码作者身份认证(CAA)支持软件取证、抄袭检测和知识产权保护。然而,现有监督CAA方法面临训练数据稀缺和封闭世界假设的限制:它们需要足够的标记代码,使得在低数据情况下训练困难,且在开放世界测试对中预测不可靠。大语言模型消除了任务特定的训练,但直接提示依赖昂贵的专家设计提示,容易在复杂异构代码对中产生幻觉,且很少产生可审计的证据轨迹。我们提出MACAA,一种基于信念修正的多智能体框架,用于无训练代码作者身份验证。MACAA包括一个协调器和四个专家代理,分析布局、词汇、语法和编程模式证据。协调器收集专家信号进行扩展,通过收缩剔除不可靠的证据,通过修正解决冲突以保持信念一致性,取代直接LLM判断,采用可审计的假设修正。MACAA在同语言基准上实现89.15%的F1分数,在混合跨语言对上实现80.00%的分数,在同语言和跨语言评估中均优于基线。

英文摘要:

Code authorship attribution (CAA) supports software forensics, plagiarism detection, and intellectual property protection. However, existing supervised CAA approaches suffer from scarce training data and closed-world assumptions: they require sufficient labeled code from fixed candidate-author sets, making training difficult in low-data cases and predictions unreliable for open-world test pairs with unseen samples, or heterogeneous code pairs. Large language models remove task-specific training, but direct prompting depends on costly expert-designed prompts, can hallucinate over complex heterogeneous code pairs, and rarely yields auditable evidence traces. We propose MACAA, a belief-revision-based multi-agent framework for training-free code authorship verification. MACAA comprises a Coordinator and four Expert Agents analyzing layout, lexical, syntactic, and programming-pattern evidence. The Coordinator gathers expert signals for expansion, discounts unreliable evidence through contraction, and resolves conflicts through revision to preserve belief consistency, replacing direct LLM judgment with auditable hypothesis refinement. MACAA achieves 89.15\% F1 on same-language benchmarks and 80.00\% on mixed cross-language pairs, outperforming the baselines overall in both same-language and cross-language evaluations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑