Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Question Answering on CovidQA (Accuracy)
Loading...
67.59
Accuracy
SFT
18.086
30.938
43.79
56.642
Mar 24, 2026
Accuracy
Updated 25d ago
Evaluation Results
Method
Method
Links
Accuracy
SFT
Backbone=Llama3.1 8B
2026.03
67.59
SFT
Backbone=Mistral 7B
2026.03
65.97
SFT
Backbone=Qwen3 1.7B
2026.03
63.15
CoA
Backbone=Llama3.1 8B
2026.03
62.14
CoA
Backbone=Qwen3 1.7B
2026.03
60.87
sudoLM
Backbone=Mistral 7B
2026.03
59.04
CoA
Backbone=Mistral 7B
2026.03
57.94
sudoLM
Backbone=Qwen3 1.7B
2026.03
56.41
Base
Backbone=Mistral 7B
2026.03
52.12
PermLM
Backbone=Mistral 7B
2026.03
48.81
Base
Backbone=Qwen3 1.7B
2026.03
48.4
Base
Backbone=Llama3.1 8B
2026.03
45.1
PermLM
Backbone=Llama3.1 8B
2026.03
43.08
PermLM
Backbone=Qwen3 1.7B
2026.03
27.5
sudoLM
Backbone=Llama3.1 8B
2026.03
19.99
Feedback
Search any
task
Search any
task