Entity Inference on ENTITY INFERENCES (test)
66.5Target AccuracyFinetuning on de + Te (full)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Finetuning on de + Te (full)Backbone=GPT-Neo 1.3B2023.06 | 66.5 | 28.8 | |
| Context Distillation (Mg = GPT3.5)Backbone=GPT-Neo 1.3B, Generator=GPT-3.52023.06 | 65.9 | 32.5 | |
| Context Distillation (Mg = GPT3.5)Backbone=GPT2-XL, Generator=GPT-3.52023.06 | 65.3 | 28.7 | |
| Prepend Def.Backbone=GPT2-XL2023.06 | 64.1 | 32.9 | |
| Finetuning on de (full)Backbone=GPT2-XL2023.06 | 62.9 | 24.1 | |
| Context Distillation (Mg = Ms)Backbone=GPT-Neo 1.3B, Generator=Self2023.06 | 61.8 | 32.6 | |
| Prepend Def.Backbone=GPT-Neo 1.3B2023.06 | 60 | 34.1 | |
| Finetuning on de + Te (full)Backbone=GPT2-XL2023.06 | 59.4 | 33.8 | |
| Context Distillation (Mg = Ms)Backbone=GPT2-XL, Generator=Self2023.06 | 58.2 | 31.4 | |
| Finetuning on de (full)Backbone=GPT-Neo 1.3B2023.06 | 57.7 | 18.3 | |
| Finetuning on de (last only)Backbone=GPT-Neo 1.3B2023.06 | 48.8 | 16.4 | |
| Finetuning on de (last only)Backbone=GPT2-XL2023.06 | 46.5 | 35.4 | |
| MENDBackbone=GPT-Neo 1.3B2023.06 | 41.8 | 34.4 | |
| Prepend Random Def.Backbone=GPT-Neo 1.3B2023.06 | 27.7 | 34.1 | |
| Prepend Random Def.Backbone=GPT2-XL2023.06 | 26.5 | 32.9 |