arXiv cs.CL· Mohammad Reza Modarres, Armin Tourajmehr, Artin Tavasoli, Yadollah Yaghoobzadeh, Mohammad Taher Pilehvar·· 12 小时前AI 评分38
每个模型都留下痕迹：多分辨率 LLM 指纹识别研究
Every Model Leaves a Trace: Multi-Resolution LLM Fingerprinting
AI 导读
研究提出“指纹分辨率”框架，从模型分辨率（跨家族、蒸馏及适配变体）和信号分辨率（行为、文体、词汇、语义）两个维度考察 LLM 生成文本的可追溯性。结果显示指纹可超越家族差异，在蒸馏模型乃至同一基座模型的 LoRA 适配版本中仍可检测；表面与词汇线索被改写后归因减弱，但仍有大量模型特异信号残留，且这些信号可跨提示分布与语言迁移，语义表征跨语言最稳健。
正文 · 原文
Abstract:Large language models leave model-specific traces in their generated text, but it remains unclear how finely these fingerprints can distinguish related models and which signals remain informative as obvious cues are reduced. We study this through fingerprint resolution, considering both model resolution, from different families to distilled and adapted variants, and signal resolution across behavioral, stylometric, lexical, and semantic evidence. We progressively suppress surface cues, paraphrase outputs, vary the amount of available text, and test generalization across prompt distributions and languages. Our results show that fingerprints persist beyond broad family differences, remaining detectable in distilled models and even LoRA adaptations of the same base model. Although attribution weakens as surface and lexical cues are altered, substantial model-specific signal remains. These signals transfer across prompt sources and languages, with semantic representations proving most robust across languages. These findings show that model identity is recoverable at multiple resolutions, but that recoverability depends on model similarity and the evidence available in the text.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.28434 [cs.CL] |
| (or arXiv:2610.28434v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.28434 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Artin Tavasoli [view email]
[v1]
Tue, 20 Oct 2026 16:00:34 UTC (198 KB)
来源：arXiv cs.CL · arxiv.org