arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2610.19583cs.AIcs.CLcs.LG
收藏

文本背后是谁?鲁棒性模型作者归属

Who's Behind the Text? Robust Model Authorship Attribution

Korea Advanced Institute of Science and Technology(韩国科学技术院)

Minseok Ko, Hyunseok Lee, Woomin Song, Beomjun Kim, Jinwoo Shin, Kunwoong Kim

AI总结:

针对现有模型作者归属(MAA)方法在分布偏移下性能下降的问题,提出ATLAS方法,结合MAABench基准验证其在OOD场景下的性能优于基线,还可用于识别蒸馏源模型。

摘要:

随着大型语言模型(LLM)越来越多地用于生成文本,其滥用已成为日益严重的社会问题,涉及学术不端、模型蒸馏和网络攻击等方面。因此,模型作者归属(MAA),即识别给定文本的源模型,对于促进LLM的负责任使用和提升AI安全性具有重要意义。在实践中,归属系统必须在训练期间观察到的分布之外仍保持可靠,它们可能会遇到来自未见领域和写作场景的文本,以及系统训练时尚未推出的模型家族未来版本的生成文本。然而,现有的MAA方法在这种分布偏移下往往会出现显著的性能下降。为应对这一挑战,我们提出了基于似然的源归属作者追踪方法(ATLAS),该方法通过源条件文本建模而非直接的源预测来构建MAA问题。ATLAS学习每个源家族生成文本的方式,并通过比较给定文本在每个学习到的源条件下的似然来归属其源。我们对似然证据进行token级别的裁剪聚合,以避免单个token主导决策。我们还引入了MAABench,这是一个广泛的MAA基准,用于评估领域和版本等实际分布外(OOD)场景。实验表明,ATLAS在MAABench和OpenTuringBench上均始终优于现有基线。特别是,当MAABench上的生成领域和模型版本均发生变化时,ATLAS达到74.27%的准确率,比最强基线高出9.00个百分点。除MAA外,ATLAS还可通过测量目标LLM最接近哪个源模型来帮助识别蒸馏源模型,我们的扩展实验显示,ATLAS可恢复我们评估的每个蒸馏模型的已知教师模型。

英文摘要:

As large language models (LLMs) are increasingly used to generate text, their misuse has become a growing societal problem, with implications for academic misconduct, model distillation, and cyberattacks. For these reasons, Model Authorship Attribution (MAA), which identifies the source model of a given text, is important for promoting the responsible use of LLMs and improving AI safety. In practice, attribution systems must remain reliable beyond the distributions observed during training. They may encounter texts from unseen domains and writing settings, as well as generations from future versions of model families that were not available when the system was trained. However, existing MAA methods often exhibit substantial performance degradation under such distribution shifts. To address this challenge, we propose Authorship Tracing via Likelihood-based Attribution of Sources (ATLAS), which formulates MAA through source-conditioned text modeling rather than direct source prediction. ATLAS learns how text from each source family is generated and attributes a given text by comparing its likelihood under each learned source condition. We aggregate this likelihood evidence at the token level with clipping so that no single token dominates the decision. We further introduce MAABench, a broad MAA benchmark for evaluating practical OOD settings such as domain and version. Our experiments show that ATLAS consistently outperforms existing baselines across MAABench and OpenTuringBench. In particular, when both the generation domain and the model version change on MAABench, ATLAS achieves 74.27% accuracy, 9.00pp higher than the strongest baseline. Beyond MAA, ATLAS can also help identify distillation source models by measuring which source model a target LLM most closely resembles. Our extended experiments show that ATLAS recovers the known teacher of every distilled model we evaluate.

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。

↑