Review Generation Evaluation on Amazon Product Reviews Third-party (test)
56HelpfulnessInterview-based Review Generation System
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Interview-based Review Generation SystemComparison Context=Comparison vs. Human2026.03 | 56 | 38 | 57 | 34 | 48 | 47 | 55 | 52 | |
| Interview-based Review Generation SystemComparison Context=Comparison vs. Baseline2026.03 | 49 | 38.5 | 47.9 | 61.5 | 50 | 52.1 | 54.2 | 44.8 | |
| BaselineComparison Context=Comparison vs. Ours2026.03 | 38.5 | 28.1 | 45.8 | 21.9 | 34.4 | 35.4 | 35.4 | 37.5 | |
| HumanComparison Context=Comparison vs. Ours2026.03 | 38 | 47 | 37 | 57 | 37 | 43 | 40 | 41 |