Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Science Question Answering on AI2SciE (test)
Loading...
95.8
Accuracy
Few-shot*
32.152
48.676
65.2
81.724
Sep 26, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Few-shot*
Backbone=Qwen2.5-7B, F...
2025.09
95.8
ICR
Backbone=Qwen2.5-7B
2025.09
93.2
M²IV
Backbone=Qwen2.5-7B
2025.09
92.4
I2CL
Backbone=Qwen2.5-7B
2025.09
91.8
LIVE
Backbone=Qwen2.5-7B
2025.09
91.8
Zero-shot
Backbone=Qwen2.5-7B
2025.09
90.4
Few-shot*
Backbone=Llama2-7B, Fe...
2025.09
45.4
ICR
Backbone=Llama2-7B
2025.09
37.2
M²IV
Backbone=Llama2-7B
2025.09
35.2
I2CL
Backbone=Llama2-7B
2025.09
35
Zero-shot
Backbone=Llama2-7B
2025.09
34.8
LIVE
Backbone=Llama2-7B
2025.09
34.6
Feedback
Search any
task
Search any
task