jinzhuoran/RWKU
RWKU: Benchmarking Real-World Knowledge Unlearning for Large Language Models. NeurIPS 2024
Python
Issues
- 2
error when running DPO
#6 opened by Wu-Zongyu - 23
Cannot replicate baseline results with LLama and Phi for Target unlearning
#3 opened by stefanvasilev - 3
code for adversary attack
#5 opened by pkulium - 2
About the calculation of metrics
#4 opened by LetheSec - 6
High Computation Cost
#2 opened by zhmzm - 1
impressive work!
#1 opened by hzy312