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

Knowledge-intensive Question Answering on GPQA (Speedup)

3.11Speedup

FailFast

1.17561.67782.182.6822Dec 23, 2025
Updated 3mo ago

Evaluation Results

MethodLinks
2025.12
3.11
2025.12
2.54
2025.12
2.42
2025.12
2.42
2025.12
2.26
2025.12
2.04
2025.12
1.98
2025.12
1.89
2025.12
1.86
2025.12
1.71
2025.12
1.34
2025.12
1.25