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 2302.06716cs.LGcs.CLcs.CR
收藏

机器学习模型归因挑战

Machine Learning Model Attribution Challenge

MITRE Corporation(MITRE公司);Plaintext Group(明文集团);Robust Intelligence(鲁棒智能公司);Hugging Face(抱抱脸)

Elizabeth Merkhofer, Deepesh Chaudhari, Hyrum S. Anderson, Keith Manville, Lily Wong, João Gante

摘要:

我们提出了机器学习模型归因挑战的结果。微调的机器学习模型可以从没有明显属性特征的其他训练模型导出。在这个挑战中,参与者只使用模型的文本输出来识别一组匿名、微调的大语言模型(LLM)的基础模型。参赛者的目标是正确地分配最微调的模型,如果参赛者的解决方案对微调模型的API的调用较少,则关系会被打破。最成功的方法是手动的,因为参与者观察到模型输出之间的相似性,并根据基础模型的公共文档开发了归因分析,尽管一些团队也提交了自动化的统计解决方案。

英文摘要:

We present the findings of the Machine Learning Model Attribution Challenge. Fine-tuned machine learning models may derive from other trained models without obvious attribution characteristics. In this challenge, participants identify the publicly-available base models that underlie a set of anonymous, fine-tuned large language models (LLMs) using only textual output of the models. Contestants aim to correctly attribute the most fine-tuned models, with ties broken in the favor of contestants whose solutions use fewer calls to the fine-tuned models' API. The most successful approaches were manual, as participants observed similarities between model outputs and developed attribution heuristics based on public documentation of the base models, though several teams also submitted automated, statistical solutions.

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。

↑