Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Paraphrase Detection on CrossFit Para (test)
Loading...
66.1
Accuracy
ABMLL
54.66
57.63
60.6
63.57
Aug 19, 2025
Accuracy
ECE
Updated 15d ago
Evaluation Results
Method
Method
Links
Accuracy
ECE
ABMLL
Model=QWEN2
2025.08
66.1
0.364
Reptile
Model=QWEN2
2025.08
63.4
0.373
Regular LoRA
Model=QWEN2
2025.08
63.3
0.423
Struct. LoRA
Model=QWEN2
2025.08
62.2
0.401
Reptile
Model=LLAMA3
2025.08
61.8
0.404
ABMLL
Model=LLAMA3
2025.08
61.6
0.413
Regular LoRA
Model=LLAMA3
2025.08
59.9
0.433
Pretrained
Model=QWEN2
2025.08
57.1
0.428
Pretrained
Model=LLAMA3
2025.08
57
0.43
Struct. LoRA
Model=LLAMA3
2025.08
55.1
0.477
Feedback
Search any
task
Search any
task