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

Closed-ended reasoning on GPQA Diamond (test)

65.9Accuracy

Majority Voting (MV)

Updated 1mo ago

Evaluation Results

Method	Links
Majority Voting (MV) 2025.12		65.9
Majority Voting (MV) 2025.12		65.2
THINKMERGE (DirectMerge (A)) 2025.12		64.1
Majority Voting (MV) 2025.12		64.1
THINKMERGE (Ours (C)) 2025.12		64.1
THINKMERGE (DirectMerge (A)) 2025.12		64
Majority Voting (MV) 2025.12		64
Majority Voting (MV) 2025.12		63.9
THINKMERGE (Ours (D)) 2025.12		63.8
THINKMERGE (Ours (D)) 2025.12		63.7
THINKMERGE (Ours (D)) 2025.12		63.7
Majority Voting (MV) 2025.12		63.4
Majority Voting (MV) 2025.12		63
THINKMERGE (Trimming (B)) 2025.12		63
THINKMERGE (Trimming (B)) 2025.12		62.7
THINKMERGE (Ours (C)) 2025.12		62.4
THINKMERGE (Ours (C)) 2025.12		62.1
THINKMERGE (DirectMerge (A)) 2025.12		61.6
THINKMERGE (Trimming (B)) 2025.12		61.4
Majority Voting (MV) 2025.12		60.8
Majority Voting (MV) 2025.12		60.8
Majority Voting (MV) 2025.12		55.5
THINKMERGE (Ours (D)) 2025.12		53.8
Majority Voting (MV) 2025.12		53.7
Majority Voting (MV) 2025.12		53.3
Majority Voting (MV) 2025.12		53.3
Majority Voting (MV) 2025.12		52.9
THINKMERGE (Ours (D)) 2025.12		52.6
Majority Voting (MV) 2025.12		52.5
Majority Voting (MV) 2025.12		52.5
THINKMERGE (Ours (C)) 2025.12		52.4
THINKMERGE (DirectMerge (A)) 2025.12		52.2
THINKMERGE (Trimming (B)) 2025.12		52.2
Majority Voting (MV) 2025.12		52
THINKMERGE (Ours (D)) 2025.12		51.8
THINKMERGE (DirectMerge (A)) 2025.12		51.6
Majority Voting (MV) 2025.12		51.4
THINKMERGE (Trimming (B)) 2025.12		51.3
THINKMERGE (Ours (C)) 2025.12		51.3
Majority Voting (MV) 2025.12		51.2
THINKMERGE (DirectMerge (A)) 2025.12		50.3
THINKMERGE (Ours (C)) 2025.12		50.3
Majority Voting (MV) 2025.12		50.3
Majority Voting (MV) 2025.12		50.3
Majority Voting (MV) 2025.12		50.2
THINKMERGE (Ours (C)) 2025.12		50.2
THINKMERGE (Trimming (B)) 2025.12		50
THINKMERGE (DirectMerge (A)) 2025.12		50
THINKMERGE (Trimming (B)) 2025.12		50
Majority Voting (MV) 2025.12		49.4
Majority Voting (MV) 2025.12		49.2
THINKMERGE (DirectMerge (A)) 2025.12		49.2
THINKMERGE (Trimming (B)) 2025.12		49.2
THINKMERGE (Ours (C)) 2025.12		49
THINKMERGE (Trimming (B)) 2025.12		49
THINKMERGE (DirectMerge (A)) 2025.12		48.9
THINKMERGE (Ours (C)) 2025.12		48.9
Majority Voting (MV) 2025.12		47.4
THINKMERGE (Ours (D)) 2025.12		47.4
THINKMERGE (Ours (D)) 2025.12		47.2
Majority Voting (MV) 2025.12		47.1
Majority Voting (MV) 2025.12		44.9
THINKMERGE (Ours (D)) 2025.12		43.8