List your company

Science Reasoning on GPQA diamond (Acc., Tokens)

72.5AccuracySC
38.80447.55256.365.048May 14, 2026

Evaluation results

MethodLinks
SCBackbone=DeepSeek-R1-0528-Qwen3-8B, Voting size budget=5122026.0572.599.2
ACBackbone=DeepSeek-R1-0528-Qwen3-8B, Voting size budget=5122026.0572.438
D-HBackbone=DeepSeek-R1-0528-Qwen3-8B, Voting size budget=5122026.0572.469.1
DDCBackbone=Qwen3-32B, Voting size budget=5122026.0572.214.1
D-LBackbone=DeepSeek-R1-0528-Qwen3-8B, Voting size budget=5122026.0571.734.6
SCBackbone=Qwen3-32B, Voting size budget=5122026.0571.174.4
D-LBackbone=Qwen3-32B, Voting size budget=5122026.0570.932.1
ACBackbone=Qwen3-32B, Voting size budget=5122026.0570.123.7
D-HBackbone=Qwen3-32B, Voting size budget=5122026.057041.6
DDCBackbone=DeepSeek-R1-0528-Qwen3-8B, Voting size budget=5122026.0569.613.9
D-LBackbone=Qwen3-8B, Voting size budget=5122026.0565.233.1
D-HBackbone=Qwen3-8B, Voting size budget=5122026.0563.849.4
SCBackbone=Qwen3-8B, Voting size budget=5122026.0563.774.7
DDCBackbone=Qwen3-8B, Voting size budget=5122026.0563.613.7
ACBackbone=Qwen3-8B, Voting size budget=5122026.0563.324.8
DDCBackbone=Qwen3-4B, Voting size budget=5122026.0559.113.7
D-LBackbone=Qwen3-4B, Voting size budget=5122026.055948.2
D-HBackbone=Qwen3-4B, Voting size budget=5122026.0548.750.6
SCBackbone=Qwen3-4B, Voting size budget=5122026.0548.580.4
ACBackbone=Qwen3-4B, Voting size budget=5122026.0548.523.7
DDCBackbone=Qwen3-1.7B, Voting size budget=5122026.0542.920.6
D-LBackbone=Qwen3-1.7B, Voting size budget=5122026.0540.4126.6
D-HBackbone=Qwen3-1.7B, Voting size budget=5122026.0540.3166.5
SCBackbone=Qwen3-1.7B, Voting size budget=5122026.0540.2186.4
ACBackbone=Qwen3-1.7B, Voting size budget=5122026.0540.169.7