Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Open-ended Visual Question Answering on Covid19 (Accuracy)
Loading...
49.43
Accuracy
T2R-R2
14.3092
23.4271
32.545
41.6629
Jun 24, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
T2R-R2
Scale=4B
2026.06
49.43
T2R-R3
Scale=4B
2026.06
41.37
T2R-R1
Scale=4B
2026.06
40.56
T2R-R3
Scale=2B
2026.06
40.37
T2R-R1
Scale=2B
2026.06
39.43
Frozen Teacher
Scale=2B
2026.06
31.88
T2R-R2
Scale=2B
2026.06
30.56
Judge Reward
Scale=4B
2026.06
29.24
RLVR Only
Scale=4B
2026.06
28.86
Frozen Teacher
Scale=4B
2026.06
27.54
Base Model
Scale=4B
2026.06
26.22
Base Model
Scale=2B
2026.06
22.64
RLVR Only
Scale=2B
2026.06
21.69
Judge Reward
Scale=2B
2026.06
15.66
Feedback
Search any
task
Search any
task