Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-turn dialogue routing on ShareGPT-LF Qwen Series Set cross-domain (legal and financial)
Loading...
91.46
Success Rate (SR)
MCTS Router
79.7912
82.8206
85.85
88.8794
Apr 14, 2026
Success Rate (SR)
Average Turns (AT)
Updated 4d ago
Evaluation Results
Method
Method
Links
Success Rate (SR)
Average Turns (AT)
MCTS Router
Candidate Set=Qwen Ser...
2026.04
91.46
4.5
Greedy Router
Candidate Set=Qwen Ser...
2026.04
87.18
4.53
DialRouter
Candidate Set=Qwen Ser...
2026.04
83.69
4.64
Qwen3-14B
Candidate Set=Qwen Ser...
2026.04
81.74
4.65
KNN Router
Candidate Set=Qwen Ser...
2026.04
81.64
4.56
Avengers
Candidate Set=Qwen Ser...
2026.04
81.3
4.67
Random
Candidate Set=Qwen Ser...
2026.04
81.03
4.67
Qwen3-8B
Candidate Set=Qwen Ser...
2026.04
80.66
4.68
Qwen3-4B
Candidate Set=Qwen Ser...
2026.04
80.24
4.8
Feedback
Search any
task
Search any
task