Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Narrow finetuning on Health
Loading...
83.9
EM-F1
Data interleaving
43.132
53.716
64.3
74.884
Jun 4, 2026
EM-F1
Utilization Change (ΔUtil.)
Updated 1mo ago
Evaluation Results
Method
Method
Links
EM-F1
Utilization Change (ΔUtil.)
Data interleaving
Model=Qwen-2.5-32B
2026.06
83.9
-0.5
TReFT
Model=Qwen-2.5-32B
2026.06
83
-0.3
TReFT
Model=Llama-3.1-8B
2026.06
80.1
0.5
TReFT
Model=Qwen-2.5-7B
2026.06
79.9
0.2
Data interleaving
Model=Qwen-2.5-7B
2026.06
76.2
-0.4
Data interleaving
Model=Llama-3.1-8B
2026.06
74.5
0.2
TReFT
Model=GPT-OSS-20B
2026.06
69.6
-1.58
SFT
Model=Qwen-2.5-32B
2026.06
61.1
-1.3
SFT
Model=GPT-OSS-20B
2026.06
54
-3.21
SFT
Model=Llama-3.1-8B
2026.06
53.4
0.1
Data interleaving
Model=GPT-OSS-20B
2026.06
53.4
-1.74
SFT
Model=Qwen-2.5-7B
2026.06
44.7
-1.8
Feedback
Search any
task
Search any
task