Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Theorem Question Answering on ThmQA
Loading...
26.12
Pass@1
Teacher
-0.0568
6.7391
13.535
20.3309
May 25, 2026
Pass@1
Updated 7d ago
Evaluation Results
Method
Method
Links
Pass@1
Teacher
Model=Gemma3-4B
2026.05
26.12
Bridge-Garden (Skew FKL)
Distillation Family=Di...
2026.05
5.24
Skew FKL
Distillation Family=Di...
2026.05
5.01
Bridge-Garden (Reverse KL)
Distillation Family=KL...
2026.05
4.57
Bridge-Garden (α-β divergence)
Distillation Family=Un...
2026.05
4.54
α-β divergence
Distillation Family=Un...
2026.05
4.34
Forward KL
Distillation Family=KL...
2026.05
4.25
Bridge-Garden (Forward KL)
Distillation Family=KL...
2026.05
4.22
Bridge-Garden (Total Variation)
Distillation Family=Di...
2026.05
4
Bridge-Garden (Skew RKL)
Distillation Family=Di...
2026.05
3.41
Skew RKL
Distillation Family=Di...
2026.05
3.38
Hard KD
Teacher Model=Gemma3-4...
2026.05
3.15
Total Variation
Distillation Family=Di...
2026.05
1.29
Reverse KL
Distillation Family=KL...
2026.05
1.26
Student (No Distill)
Model=Gemma3-1B
2026.05
0.95
Feedback
Search any
task
Search any
task