List your company

Multi-hop Question Answering on MuSiQue (pass@1, pass@2)

69.4Pass@1EXG-SE-Rev
15.3229.3643.457.44May 18, 2026

Evaluation results

MethodLinks
EXG-SE-RevModel=Qwen-Plus2026.0569.499
EXG-ReflexionModel=Qwen-Plus2026.0565.294.6
EXG-SEModel=Qwen-Plus2026.0565.285.4
EXGModel=Qwen-Plus2026.0560.867.2
SE-Agent-RevModel=Qwen-Plus2026.0556.897
EXG-SE-RevModel=Qwen3-14B2026.055689.6
SE-AgentModel=Qwen-Plus2026.0555.284
ReflexionModel=Qwen-Plus2026.0554.294
EXG-SE-RevModel=Qwen-Max2026.0551.892.4
EXG-SEModel=Qwen3-14B2026.054973.6
EXG-SEModel=Qwen-Max2026.0548.692.6
EXG-ReflexionModel=Qwen3-14B2026.054874.8
EXG-ReflexionModel=Qwen-Max2026.0547.876.2
EXGModel=Qwen-Max2026.0544.651.6
SE-Agent-RevModel=Qwen3-14B2026.0538.283.6
SE-Agent-RevModel=Qwen-Max2026.053788.2
SE-AgentModel=Qwen3-14B2026.0535.670
SE-AgentModel=Qwen-Max2026.0534.874.6
ReflexionModel=Qwen-Max2026.0534.669.2
EXGModel=Qwen3-14B2026.053240.6
ReflexionModel=Qwen3-14B2026.0517.432.6