Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Machine Unlearning on TOFU 1.0 (Real Author)
Loading...
93
ROUGE-L
LLAMA2-7B-TOFU
59.72
68.36
77
85.64
Oct 23, 2024
ROUGE-L
Probability
Truth Ratio
Updated 4d ago
Evaluation Results
Method
Method
Links
ROUGE-L
Probability
Truth Ratio
LLAMA2-7B-TOFU
Model Type=Base (No Un...
2024.10
93
0.45
58
LLAMA2-13B-TOFU
Model Type=Base (No Un...
2024.10
89
0.51
67
LoRA
Base Model=LLAMA2-7B-TOFU
2024.10
88
0.51
68
LoRA
Base Model=LLAMA2-13B-...
2024.10
87
0.54
70
CMC
Base Model=LLAMA2-7B-TOFU
2024.10
84
0.43
58
MISTRAL-7B-TOFU
Model Type=Base (No Un...
2024.10
84
0.61
75
LoRA
Base Model=MISTRAL-7B-...
2024.10
79
0.57
71
CMC
Base Model=LLAMA2-13B-...
2024.10
77
0.49
65
CMC
Base Model=MISTRAL-7B-...
2024.10
73
0.62
75
δ-UNLEARNING
Base Model=LLAMA2-13B-...
2024.10
61
0.36
46
Feedback
Search any
task
Search any
task