Natural Language Explanation Generation on ComVE few-shot 60-shot
68.03AccuracySPARSEFIT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SPARSEFITFine-tuning Strategy=Att.Q+LN, Backbone=Llama 2-7B, Percentage Parameters=7.97%, Few-shot setting=60-shot2023.05 | 68.03 | 58.67 | |
| AdaLoRAFine-tuning Strategy=PEFT, Backbone=Llama 2-7B, Percentage Parameters=0.30%, Few-shot setting=60-shot2023.05 | 64.23 | 56.16 | |
| Full Fine-tuningFine-tuning Strategy=Baseline, Backbone=Llama 2-7B, Percentage Parameters=100%, Few-shot setting=60-shot2023.05 | 63.71 | 55.93 |