TOFU_Task_of_Fictitious_Unlearning

TOFU: Task of Fictitious Unlearning

The TOFU dataset serves as a benchmark for evaluating unlearning performance of large language models on realistic tasks.

MIT
10K – 100K
Question Answering
English
by @AIOZAI
211

Last updated: 8 months ago


Sign in to see dataset files

orCreate an account