Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Narrow finetuning on Legal
Loading...
85.9
EM-F1
TReFT
52.204
60.952
69.7
78.448
Jun 4, 2026
EM-F1
Utilization Delta (ΔUtil)
Updated 1mo ago
Evaluation Results
Method
Method
Links
EM-F1
Utilization Delta (ΔUtil)
TReFT
Model=Qwen-2.5-32B
2026.06
85.9
-0.3
Data interleaving
Model=Qwen-2.5-32B
2026.06
81.7
-1.2
TReFT
Model=GPT-OSS-20B
2026.06
79.2
-1.85
TReFT
Model=Llama-3.1-8B
2026.06
78.4
-0.2
TReFT
Model=Qwen-2.5-7B
2026.06
77.7
-0.7
Data interleaving
Model=GPT-OSS-20B
2026.06
75.1
-2.25
Data interleaving
Model=Qwen-2.5-7B
2026.06
74.9
-0.7
Data interleaving
Model=Llama-3.1-8B
2026.06
73
-0.9
SFT
Model=Llama-3.1-8B
2026.06
61.4
-0.9
SFT
Model=GPT-OSS-20B
2026.06
61.3
-3.3
SFT
Model=Qwen-2.5-32B
2026.06
53.7
-2.4
SFT
Model=Qwen-2.5-7B
2026.06
53.5
-1.1
Feedback
Search any
task
Search any
task