E-commerce Price Optimization on Tao Factory Online A/B Test 7-Day Horizon
6.92MARAIGP
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| AIGPMode=SFT+DPO2026.06 | 6.92 | 11.04 | 8.23 | |
| Qwen3-235B-A22BBackbone=Qwen3-235B-A22B2026.06 | 5.69 | 8.67 | 8.52 | |
| AIGPMode=SFT-only2026.06 | 5.59 | 4.82 | 8 | |
| RL-DT2026.06 | 1.33 | 4.02 | 5.71 | |
| Qwen3-30B-A3BBackbone=Qwen3-30B-A3B2026.06 | 0.88 | 2.22 | 2.29 |