Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Knowledge Unlearning on MQUAKE Few-shot
Loading...
29.16
Efficiency
Base
0.4352
7.8926
15.35
22.8074
May 16, 2026
Efficiency
Perplexity
Updated 14d ago
Evaluation Results
Method
Method
Links
Efficiency
Perplexity
Base
Model=Qwen-3
2026.05
29.16
11.21
ROME
Model=Qwen-3
2026.05
28.92
11.28
MEMIT
Model=Qwen-3
2026.05
25.55
11.27
FT
Model=Qwen-3
2026.05
24.91
11.55
AlphaEdit
Model=Qwen-3
2026.05
24.62
11.29
ZeroUnlearn
Model=Qwen-3
2026.05
22.98
11.29
GA
Model=Qwen-3
2026.05
1.54
455.19
Feedback
Search any
task
Search any
task
