List your company

Multi-turn dialogue on JDDC

88.42Success Rate (SR)MCTS Router
46.45657.350568.24579.1395Apr 14, 2026

Evaluation results

MethodLinks
MCTS RouterCandidate LLM Set=Qwen Series Set2026.0488.423.55
MCTS RouterCandidate LLM Set=Llama Series Set2026.0485.283.86
Greedy RouterCandidate LLM Set=Qwen Series Set2026.0483.63.54
DialRouterCandidate LLM Set=Qwen Series Set2026.0483.453.47
Greedy RouterCandidate LLM Set=Llama Series Set2026.0480.833.73
DialRouterCandidate LLM Set=Llama Series Set2026.0480.163.9
KNN RouterCandidate LLM Set=Qwen Series Set2026.0478.443.32
AvengersCandidate LLM Set=Qwen Series Set2026.0478.393.57
Qwen3-14BCandidate LLM Set=Qwen Series Set2026.0477.193.36
RouterDCCandidate LLM Set=Qwen Series Set2026.0477.193.36
RandomCandidate LLM Set=Qwen Series Set2026.0476.63.52
MF RouterCandidate LLM Set=Qwen Series Set2026.0476.593.73
Prompt LLMCandidate LLM Set=Qwen Series Set2026.0476.493.5
Qwen3-8BCandidate LLM Set=Qwen Series Set2026.0476.143.5
KNN RouterCandidate LLM Set=Llama Series Set2026.0474.823.72
AvengersCandidate LLM Set=Llama Series Set2026.0473.993.7
RouterDCCandidate LLM Set=Llama Series Set2026.0473.783.6
Qwen3-4BCandidate LLM Set=Qwen Series Set2026.0473.453.82
Llama3.1-8B-InstructCandidate LLM Set=Llama Series Set2026.0473.213.59
Prompt LLMCandidate LLM Set=Llama Series Set2026.0472.493.91
MF RouterCandidate LLM Set=Llama Series Set2026.0472.063.66
Llama3.2-3B-InstructCandidate LLM Set=Llama Series Set2026.0470.653.94
RandomCandidate LLM Set=Llama Series Set2026.0468.974
Llama3.2-1B-InstructCandidate LLM Set=Llama Series Set2026.0448.074.43