POI Attribute Acquisition (Task-oriented Dialogue) on DuIVRS logs Drobust 1.0 (test)
76.22Completion Rate (CR)DuIVRS-2
Evaluation Results
| Method | Links | |
|---|---|---|
| DuIVRS-2LLM-S=ERNIE-Bot-tiny, LLM-L=ERNIE-Bot-turbo, Black-box LLM=ERNIE 4.02026.05 | 76.22 | |
| HybridLLMsLLM-S=Qwen2.5-1.5B, LLM-L=Qwen2.5-7B, Black-box LLM=GPT-4o2026.05 | 75.89 | |
| DuIVRS-12026.05 | 69.05 | |
| DeepSeek-V32026.05 | 64.55 | |
| GPT-4o2026.05 | 63.13 | |
| LLM-DM2026.05 | 60.98 | |
| w/o-DA2026.05 | 54.12 | |
| Direct-SFT2026.05 | 52.9 | |
| w/o-CoT2026.05 | 28.35 |