List your company

Graduate-level Question Answering on GPQA Diamond (test)

37.4AccuracyAnchor-Points
29.49631.54833.635.652Jan 7, 2026

Evaluation results

MethodLinks
Anchor-PointsOptimization Budget=Large2026.0137.4
SESS-wrepOptimization Budget=Small2026.0137.4
SESS-vlcOptimization Budget=Large2026.0136.9
RandomOptimization Budget=Large2026.0135.9
SESS-wrepOptimization Budget=Large2026.0134.8
RandomOptimization Budget=Small2026.0134.3
SESS-lcOptimization Budget=Small2026.0134.3
SESS-vlcOptimization Budget=Small2026.0134.3
IPOMPOptimization Budget=Small2026.0133.8
IPOMPOptimization Budget=Large2026.0133.8
Anchor-PointsOptimization Budget=Small2026.0133.8
SESS-lcOptimization Budget=Large2026.0132.8
SESS-repOptimization Budget=Small2026.0132.3
BaseOptimization Budget=Small2026.0129.9
BaseOptimization Budget=Large2026.0129.9
SESS-repOptimization Budget=Large2026.0129.8