Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Science Reasoning on AI2ARC
Loading...
92.92
Accuracy
D-RPC
75.3128
79.8839
84.455
89.0261
May 8, 2026
Accuracy
Updated 23d ago
Evaluation Results
Method
Method
Links
Accuracy
D-RPC
Student Model=Llama 3....
2026.05
92.92
Freeform
Student Model=Llama 3....
2026.05
92.41
D-RPC
Student Model=Qwen 3 1.7B
2026.05
88.82
CoT
Student Model=Qwen 3 1.7B
2026.05
88.78
Freeform
Student Model=Qwen 3 1.7B
2026.05
88.7
CoT
Student Model=Llama 3....
2026.05
87.92
SGFT
Student Model=Llama 3....
2026.05
86.71
DCoT
Student Model=Qwen 3 1.7B
2026.05
85.74
DCoT
Student Model=Llama 3....
2026.05
80.78
SGFT
Student Model=Qwen 3 1.7B
2026.05
75.99
Feedback
Search any
task
Search any
task