Data Free Knowledge Distillation On Squad
Metriken
Exact Match
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
| Paper Title | ||
|---|---|---|
| GOLD (T5-base) | 75.2 | GOLD: Generalized Knowledge Distillation via Out-of-Distribution-Guided Language Data Generation |
| Prompt2Model (T5-base) | 74.4 | Prompt2Model: Generating Deployable Models from Natural Language Instructions |
| ZeroGen (T5-base) | 69.4 | ZeroGen: Efficient Zero-shot Learning via Dataset Generation |
| ProGen (T5-base) | 68.1 | ProGen: Progressive Zero-shot Dataset Generation via In-context Feedback |
0 of 4 row(s) selected.