Question Answering on HotpotQA (F1 and Speedup)
27.87F1 ScoreProphet
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ProphetSchedule=Prophet2025.12 | 27.87 | 4.51 | |
| BaselineSchedule=Standard2025.12 | 27.51 | 1 | |
| Linear (7.5, 0)Schedule Type=Linear, Thresholds (τ_high, τ_low)=(7.5, 0)2025.12 | 27.48 | 3.26 | |
| Cosine (7.5, 2.5)Schedule Type=Cosine, Thresholds (τ_high, τ_low)=(7.5, 2.5)2025.12 | 27.47 | 3.14 | |
| Linear (7.5, 2.5)Schedule Type=Linear, Thresholds (τ_high, τ_low)=(7.5, 2.5)2025.12 | 27.47 | 3.17 | |
| Cosine (7.5, 0)Schedule Type=Cosine, Thresholds (τ_high, τ_low)=(7.5, 0)2025.12 | 27.46 | 3.24 | |
| Exp-k=2 (7.5, 2.5)Schedule Type=Exponential, k=2, Thresholds (τ_high, τ_low)=(7.5, 2.5)2025.12 | 26.91 | 3.23 | |
| Exp-k=2 (7.5, 0)Schedule Type=Exponential, k=2, Thresholds (τ_high, τ_low)=(7.5, 0)2025.12 | 26.71 | 3.47 | |
| Exp-k=16 (7.5, 2.5)Schedule Type=Exponential, k=16, Thresholds (τ_high, τ_low)=(7.5, 2.5)2025.12 | 26.58 | 4.33 | |
| Exp-k=16 (7.5, 0)Schedule Type=Exponential, k=16, Thresholds (τ_high, τ_low)=(7.5, 0)2025.12 | 26.2 | 5.79 |