Sentiment Classification on Amz5 (test)
97AccuracyICL-gold
Evaluation Results
| Method | Links | |
|---|---|---|
| ICL-goldPrompting Method=Direct, Backbone=GPT-3, Oracle=true2022.12 | 97 | |
| ICL-randomPrompting Method=Direct, Backbone=GPT-3, Oracle=true2022.12 | 95.4 | |
| Z-ICLPrompting Method=Direct, Backbone=GPT-32022.12 | 93 | |
| Z-ICLPrompting Method=Channel, Backbone=GPT-32022.12 | 89.1 | |
| No-demosPrompting Method=Direct, Backbone=GPT-32022.12 | 88.2 | |
| ICL-goldPrompting Method=Channel, Backbone=GPT-3, Oracle=true2022.12 | 86 | |
| ICL-randomPrompting Method=Channel, Backbone=GPT-3, Oracle=true2022.12 | 83.4 | |
| No-demosPrompting Method=Channel, Backbone=GPT-32022.12 | 77.2 | |
| Majority2022.12 | 50 | |
| ICL-goldModel=GPT-J, Prompting=Direct, Training data access=true2022.12 | 49 | |
| ICL-goldModel=GPT-NeoX, Prompting=Direct, Training data access=true2022.12 | 47 | |
| Z-ICLModel=GPT-J, Prompting=Channel, Training data access=false2022.12 | 46.5 | |
| ICL-randomModel=GPT-NeoX, Prompting=Direct, Training data access=true2022.12 | 45.6 | |
| ICL-goldModel=GPT-J, Prompting=Channel, Training data access=true2022.12 | 45.5 | |
| ICL-randomModel=GPT-J, Prompting=Channel, Training data access=true2022.12 | 44.9 | |
| Naive Z-ICLModel=GPT-J, Prompting=Channel, Training data access=false2022.12 | 41.7 | |
| ICL-goldModel=GPT-NeoX, Prompting=Channel, Training data access=true2022.12 | 41.6 | |
| Naive Z-ICLModel=GPT-NeoX, Prompting=Direct, Training data access=false2022.12 | 41.2 | |
| Z-ICLModel=GPT-NeoX, Prompting=Direct, Training data access=false2022.12 | 41.2 | |
| ICL-randomModel=GPT-J, Prompting=Direct, Training data access=true2022.12 | 41.1 | |
| ICL-randomModel=GPT-NeoX, Prompting=Channel, Training data access=true2022.12 | 39.8 | |
| Naive Z-ICLModel=GPT-J, Prompting=Direct, Training data access=false2022.12 | 39.6 | |
| Random inputsModel=GPT-NeoX, Prompting=Direct, Training data access=false2022.12 | 38.7 | |
| Z-ICLModel=GPT-J, Prompting=Direct, Training data access=false2022.12 | 38.5 | |
| Random inputsModel=GPT-J, Prompting=Channel, Training data access=false2022.12 | 38.1 | |
| Z-ICLModel=GPT-NeoX, Prompting=Channel, Training data access=false2022.12 | 37.8 | |
| Random inputsModel=GPT-J, Prompting=Direct, Training data access=false2022.12 | 37.5 | |
| Naive Z-ICLModel=GPT-NeoX, Prompting=Channel, Training data access=false2022.12 | 34.7 | |
| No-demosModel=GPT-J, Prompting=Channel, Training data access=false2022.12 | 34.4 | |
| No-demosModel=GPT-J, Prompting=Direct, Training data access=false2022.12 | 30.4 | |
| Random inputsModel=GPT-NeoX, Prompting=Channel, Training data access=false2022.12 | 27.9 | |
| No-demosModel=GPT-NeoX, Prompting=Channel, Training data access=false2022.12 | 27.5 | |
| No-demosModel=GPT-NeoX, Prompting=Direct, Training data access=false2022.12 | 20.2 | |
| MajorityTraining data access=false2022.12 | 20 |