Dialogue Planning on CraigslistBargain
2.32ATDialogXpert
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| DialogXpertBackbone=Qwen2.5 14B2026.04 | 2.32 | 97.46 | 0.4389 | |
| NRPA-GD level 2Backbone=GPT-4o mini2026.04 | 2.72 | 100 | 0.6422 | |
| DialogXpert -w/o EmotionBackbone=Qwen2.5 14B2026.04 | 2.75 | 91.36 | 0.3156 | |
| DPDP -System 2Backbone=GPT-3.5-Turbo2026.04 | 2.78 | 97.34 | 0.2728 | |
| DialogXpertBackbone=Qwen 1.8B2026.04 | 2.78 | 92.74 | 0.3791 | |
| UP-NRPA level 2Backbone=Qwen2.5 14B2026.04 | 2.88 | 100 | 0.6865 | |
| DialogXpertBackbone=Vicuna 13B2026.04 | 2.93 | 94.15 | 0.3811 | |
| UP-NRPA level 1Backbone=Qwen2.5 14B2026.04 | 3.14 | 100 | 0.5835 | |
| DialogXpert -w/o LLM-PriorBackbone=Qwen2.5 14B2026.04 | 3.31 | 91.65 | 0.3598 | |
| DialogXpert -w/o RLBackbone=Qwen2.5 14B2026.04 | 3.64 | 87.54 | 0.2952 | |
| NRPA-GD level 1Backbone=GPT-4o mini2026.04 | 3.89 | 98.94 | 0.6326 | |
| DialogXpert -w/o RLBackbone=Qwen 1.8B2026.04 | 4.69 | 77.54 | 0.3012 | |
| UP-NRPA level 2Backbone=GPT-4o mini2026.04 | 4.74 | 97.87 | 0.9069 | |
| DPDP (System 1)Backbone=GPT-3.5-Turbo2026.04 | 5.03 | 74.47 | 0.4108 | |
| DialogXpert -w/o RLBackbone=Vicuna 13B2026.04 | 5.13 | 75.61 | 0.3473 | |
| UP-NRPA level 1Backbone=GPT-4o mini2026.04 | 5.22 | 90.96 | 0.7882 | |
| PPDPP -w/o RLBackbone=Vicuna 13B2026.04 | 5.57 | 66.49 | 0.228 | |
| PPDPPBackbone=Vicuna 13B2026.04 | 5.62 | 61.17 | 0.3376 | |
| PPDPP -w/o SFTBackbone=Vicuna 13B2026.04 | 5.71 | 62.23 | 0.3354 | |
| Proactive + MI-PromptBackbone=-2026.04 | 5.74 | 56.91 | 0.268 | |
| ProactiveBackbone=-2026.04 | 5.8 | 56.38 | 0.2489 | |
| AnEBackbone=-2026.04 | 5.91 | 45.21 | 0.2608 | |
| ProCoT + MI-PromptBackbone=-2026.04 | 6.12 | 55.32 | 0.3059 | |
| ProCoTBackbone=-2026.04 | 6.22 | 53.19 | 0.2486 | |
| StandardBackbone=-2026.04 | 6.47 | 38.3 | 0.1588 | |
| ICL-AIFBackbone=-2026.04 | 6.53 | 36.17 | 0.1881 | |
| DialoGPTBackbone=GPT-22026.04 | 6.73 | 32.45 | 0.2012 |