Natural Language Explanation Generation on ComVE
70Human Evaluation ScoreSPARSEFIT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SPARSEFITBackbone=Llama-2-7B, Setting=Att.Q+LN2023.05 | 70 | — | |
| AdaLoraBackbone=Llama-2-7B2023.05 | 55.56 | — | |
| SPARSEFIT (Att.Q+LN)Backbone=T5-large, Fine-tuning Strategy=Att.Q+LN2023.05 | 40 | 0.25 | |
| Full Fine-tuningBackbone=Llama-2-7B2023.05 | 40 | — | |
| SPARSEFIT (Att.Q)Backbone=T5-large, Fine-tuning Strategy=Att.Q2023.05 | 28.89 | 0.35 | |
| AdaLoraBackbone=T5-large, Fine-tuning Strategy=AdaLoRA, Trainable Parameters (%)=1.15%2023.05 | 23.34 | 0.25 | |
| Full Fine-tuningBackbone=T5-large, Fine-tuning Strategy=Full Fine-tuning, Trainable Parameters (%)=100%2023.05 | 21.67 | 0.22 |