Sentiment Transformation on IMDB NegToPos 1.0 (test)
0.566RelevanceINJECTION
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| INJECTIONModel=LLaMA-2 13B2024.02 | 0.566 | 83.866 | 28.3 | |
| INJECTIONModel=OPT 6.7B2024.02 | 0.556 | 36.38 | 41.7 | |
| INJECTIONModel=Alpaca 7B2024.02 | 0.532 | 55.891 | 45.4 | |
| DATG-PModel=OPT 6.7B2024.02 | 0.525 | 33.768 | 29.5 | |
| DATG-PModel=Falcon 7B2024.02 | 0.513 | 48.349 | 41.4 | |
| INJECTIONModel=Phi-2 2.7B2024.02 | 0.513 | 64.785 | 40.7 | |
| CONTINUATIONModel=OPT 6.7B2024.02 | 0.51 | 23.954 | 33.3 | |
| DATG-PModel=Alpaca 7B2024.02 | 0.508 | 72.195 | 30.9 | |
| DATG-PModel=LLaMA-2 13B2024.02 | 0.508 | 60.189 | 36.5 | |
| INJECTIONModel=Falcon 7B2024.02 | 0.502 | 36.852 | 47.7 | |
| CONTINUATIONModel=Alpaca 7B2024.02 | 0.5 | 37.58 | 36.4 | |
| CONTINUATIONModel=LLaMA-2 13B2024.02 | 0.499 | 37.759 | 38.4 | |
| CONTINUATIONModel=Falcon 7B2024.02 | 0.498 | 31.599 | 35.7 | |
| PREADDModel=Falcon 7B2024.02 | 0.492 | 64.122 | 39 | |
| PREADDModel=OPT 6.7B2024.02 | 0.49 | 54.107 | 31.7 | |
| DATG-LModel=OPT 6.7B2024.02 | 0.472 | 26.634 | 42.8 | |
| CONTINUATIONModel=Phi-2 2.7B2024.02 | 0.472 | 28.844 | 39.4 | |
| DATG-PModel=Phi-2 2.7B2024.02 | 0.467 | 41.663 | 29 | |
| PREADDModel=Alpaca 7B2024.02 | 0.465 | 73.021 | 39.5 | |
| DATG-LModel=Falcon 7B2024.02 | 0.462 | 30.749 | 47.8 | |
| DATG-LModel=LLaMA-2 13B2024.02 | 0.456 | 39.382 | 46.4 | |
| DATG-LModel=Phi-2 2.7B2024.02 | 0.455 | 27.103 | 45.8 | |
| PREADDModel=LLaMA-2 13B2024.02 | 0.453 | 76.535 | 41.6 | |
| DATG-LModel=Alpaca 7B2024.02 | 0.447 | 37.295 | 46.7 | |
| PREADDModel=Phi-2 2.7B2024.02 | 0.437 | 39.458 | 47.4 | |
| FUDGEModel=OPT 6.7B2024.02 | 0.411 | 198.18 | 24.7 | |
| FUDGEModel=Phi-2 2.7B2024.02 | 0.398 | 206.452 | 31.5 | |
| FUDGEModel=Falcon 7B2024.02 | 0.397 | 193.347 | 34.7 | |
| FUDGEModel=LLaMA-2 13B2024.02 | 0.394 | 219.241 | 29.1 | |
| FUDGEModel=Alpaca 7B2024.02 | 0.392 | 208.181 | 31.8 |