尹志
(2025-04-30 15:56):
#paper doi:10.48550/arXiv.2407.20516, Machine Unlearning in Generative AI: A Survey. 很有意思的方向,应该是翻译机器遗忘吧。随着模型越做越大,如何通过对模型的处理达到可控的添加与擦除特定信息,是未来一个重要的主题,不管是从隐私保护还是模型控制的层面上
arXiv,
2024-07-30T03:26:09Z.
DOI: 10.48550/arXiv.2407.20516
Machine Unlearning in Generative AI: A Survey
翻译
Abstract:
Generative AI technologies have been deployed in many places, such as(multimodal) large language models and vision generative models. Theirremarkable performance should be attributed to massive training data andemergent reasoning abilities. However, the models would memorize and generatesensitive, biased, or dangerous information originated from the training dataespecially those from web crawl. New machine unlearning (MU) techniques arebeing developed to reduce or eliminate undesirable knowledge and its effectsfrom the models, because those that were designed for traditionalclassification tasks could not be applied for Generative AI. We offer acomprehensive survey on many things about MU in Generative AI, such as a newproblem formulation, evaluation methods, and a structured discussion on theadvantages and limitations of different kinds of MU techniques. It alsopresents several critical challenges and promising directions in MU research. Acurated list of readings can be found:https://github.com/franciscoliu/GenAI-MU-Reading.
翻译
Related Links: