arXiv cs.AI· Jinmo Han, Ju Yeon Kang, Jimin Hong, Sunghwan Ahn, Yonghyeon Jun, Nam Soo Kim·· 2 小时前AI 评分34
SetPrint：基于响应变化集合对应关系的黑盒 LLM 指纹识别方法
SetPrint: Black-Box LLM Fingerprinting via Setwise Correspondence of Response Changes
AI 导读
SetPrint 是一种免训练的黑盒 LLM 指纹识别方法，通过扰动查询并联合捕捉响应变化的集合内关系与集合间对应关系，判断目标模型是否由源模型衍生。该方法将每个响应变化相对自身集合均值表示，再比较两模型在同一扰动下的变化。在 LeaFBench 上，SetPrint 持续超越现有免训练黑盒指纹识别方法。
正文 · 原文
Abstract:Black-box LLM fingerprinting aims to determine whether a target model was derived from a source model when only the output responses to input queries are observable. One way to reveal this relationship is to perturb each query and observe the resulting responses of both models, yielding two corresponding sets of response variations. These sets provide two complementary sources of information: the relations among response changes within each set, and the correspondence of response changes between the two sets. Existing approaches, however, have not fully exploited both. We introduce \textbf{SetPrint}, a training-free method that jointly captures both within-set relations and between-set correspondence. SetPrint expresses each response change relative to the mean change of its own set, and then compares changes from the two models that were induced by the same perturbation. On LeaFBench, SetPrint consistently outperforms existing training-free black-box fingerprinting methods. Further analyses attribute these gains to both within-set relations and between-set correspondence.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.12950 [cs.AI] |
| (or arXiv:2611.12950v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.12950 arXiv-issued DOI via DataCite |
Submission history
From: Jinmo Han [view email]
[v1]
Sat, 7 Nov 2026 00:45:17 UTC (82 KB)
来源：arXiv cs.AI · arxiv.org