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

Multi-turn dialogue on ShareGPT, JDDC, and MedDG Aggregated

89.77SRavg

MCTS Router

50.04260.35670.6780.984Apr 14, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.04
89.77
2026.04
86.53
2026.04
83.94
2026.04
83.31
2026.04
79.7
2026.04
78.88
2026.04
78
2026.04
77.93
2026.04
77.64
2026.04
77.47
2026.04
77.36
2026.04
77.14
2026.04
76.92
2026.04
76.52
2026.04
75.41
2026.04
73.7
2026.04
73.6
2026.04
73.5
2026.04
72.75
2026.04
72.6
2026.04
72.34
2026.04
68.75
2026.04
68.4
2026.04
51.57