Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Multilingual Question Answering on mGPQA

32.9Accuracy

self-cons (ours)

18.23622.04325.8529.657May 31, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
32.9
2026.05
32.1
2026.05
29.4
2026.05
28.9
2026.05
28.2
2026.05
25.7
2026.05
24.9
2026.05
24.6
2026.05
24.5
2026.05
23.5
2026.05
22.8
2026.05
18.8