List your company

Scientific Verification on GPQA DIAMOND

65.15EMPRISM-MCTS
45.02650.250555.47560.6995Apr 7, 2026

Evaluation results

MethodLinks
PRISM-MCTSModel=Qwen3-30B-A3B-Instruct-25072026.0465.1566.26
PRISM-MCTSModel=GPT-4.1-mini2026.0465.0865.16
MCTS-RAGModel=GPT-4.1-mini2026.0464.6564.8
MCTS-RAGModel=Qwen3-30B-A3B-Instruct-25072026.0463.6464.26
Search-O1Model=Qwen3-30B-A3B-Instruct-25072026.0461.1153.76
ReST-MCTS*Model=GPT-4.1-mini2026.0460.6160.53
ReST-MCTS*Model=Qwen3-30B-A3B-Instruct-25072026.0458.0860.45
Search-O1Model=GPT-4.1-mini2026.0457.0757.16
ReActModel=GPT-4.1-mini2026.0455.0560.75
ReActModel=Qwen3-30B-A3B-Instruct-25072026.0448.9953.49
Zero-shotModel=GPT-4.1-mini2026.0447.5546.55
Zero-shotModel=Qwen3-30B-A3B-Instruct-25072026.0445.845.29