Multi-turn dialogue on JDDC
88.42Success Rate (SR)MCTS Router
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MCTS RouterCandidate LLM Set=Qwen Series Set2026.04 | 88.42 | 3.55 | |
| MCTS RouterCandidate LLM Set=Llama Series Set2026.04 | 85.28 | 3.86 | |
| Greedy RouterCandidate LLM Set=Qwen Series Set2026.04 | 83.6 | 3.54 | |
| DialRouterCandidate LLM Set=Qwen Series Set2026.04 | 83.45 | 3.47 | |
| Greedy RouterCandidate LLM Set=Llama Series Set2026.04 | 80.83 | 3.73 | |
| DialRouterCandidate LLM Set=Llama Series Set2026.04 | 80.16 | 3.9 | |
| KNN RouterCandidate LLM Set=Qwen Series Set2026.04 | 78.44 | 3.32 | |
| AvengersCandidate LLM Set=Qwen Series Set2026.04 | 78.39 | 3.57 | |
| Qwen3-14BCandidate LLM Set=Qwen Series Set2026.04 | 77.19 | 3.36 | |
| RouterDCCandidate LLM Set=Qwen Series Set2026.04 | 77.19 | 3.36 | |
| RandomCandidate LLM Set=Qwen Series Set2026.04 | 76.6 | 3.52 | |
| MF RouterCandidate LLM Set=Qwen Series Set2026.04 | 76.59 | 3.73 | |
| Prompt LLMCandidate LLM Set=Qwen Series Set2026.04 | 76.49 | 3.5 | |
| Qwen3-8BCandidate LLM Set=Qwen Series Set2026.04 | 76.14 | 3.5 | |
| KNN RouterCandidate LLM Set=Llama Series Set2026.04 | 74.82 | 3.72 | |
| AvengersCandidate LLM Set=Llama Series Set2026.04 | 73.99 | 3.7 | |
| RouterDCCandidate LLM Set=Llama Series Set2026.04 | 73.78 | 3.6 | |
| Qwen3-4BCandidate LLM Set=Qwen Series Set2026.04 | 73.45 | 3.82 | |
| Llama3.1-8B-InstructCandidate LLM Set=Llama Series Set2026.04 | 73.21 | 3.59 | |
| Prompt LLMCandidate LLM Set=Llama Series Set2026.04 | 72.49 | 3.91 | |
| MF RouterCandidate LLM Set=Llama Series Set2026.04 | 72.06 | 3.66 | |
| Llama3.2-3B-InstructCandidate LLM Set=Llama Series Set2026.04 | 70.65 | 3.94 | |
| RandomCandidate LLM Set=Llama Series Set2026.04 | 68.97 | 4 | |
| Llama3.2-1B-InstructCandidate LLM Set=Llama Series Set2026.04 | 48.07 | 4.43 |