Sentiment Transformation on IMDB PosToNeg 1.0 (test)
0.556RelevanceINJECTION
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| INJECTIONModel=LLaMA-2 13B2024.02 | 0.556 | 79.626 | 35.6 | |
| INJECTIONModel=OPT 6.7B2024.02 | 0.548 | 41.175 | 37.2 | |
| INJECTIONModel=Alpaca 7B2024.02 | 0.538 | 63.483 | 39.6 | |
| INJECTIONModel=Falcon 7B2024.02 | 0.516 | 34.296 | 32.8 | |
| DATG-PModel=Falcon 7B2024.02 | 0.514 | 47.28 | 32.8 | |
| CONTINUATIONModel=OPT 6.7B2024.02 | 0.513 | 25.48 | 26.9 | |
| CONTINUATIONModel=LLaMA-2 13B2024.02 | 0.51 | 41.397 | 18.8 | |
| INJECTIONModel=Phi-2 2.7B2024.02 | 0.51 | 62.835 | 36.2 | |
| DATG-PModel=Alpaca 7B2024.02 | 0.506 | 75.275 | 42.6 | |
| DATG-PModel=LLaMA-2 13B2024.02 | 0.505 | 66.427 | 41.8 | |
| CONTINUATIONModel=Alpaca 7B2024.02 | 0.502 | 39.887 | 20.3 | |
| DATG-PModel=OPT 6.7B2024.02 | 0.501 | 34.08 | 49 | |
| CONTINUATIONModel=Falcon 7B2024.02 | 0.498 | 33.714 | 20.6 | |
| PREADDModel=OPT 6.7B2024.02 | 0.48 | 50.183 | 33.1 | |
| PREADDModel=Falcon 7B2024.02 | 0.477 | 65.083 | 25.6 | |
| DATG-PModel=Phi-2 2.7B2024.02 | 0.472 | 44.139 | 46.4 | |
| PREADDModel=LLaMA-2 13B2024.02 | 0.469 | 75.418 | 23.8 | |
| CONTINUATIONModel=Phi-2 2.7B2024.02 | 0.467 | 35.489 | 18.4 | |
| DATG-LModel=OPT 6.7B2024.02 | 0.459 | 25.487 | 35.7 | |
| PREADDModel=Alpaca 7B2024.02 | 0.457 | 77.644 | 28.6 | |
| DATG-LModel=Alpaca 7B2024.02 | 0.453 | 46.061 | 33.2 | |
| DATG-LModel=LLaMA-2 13B2024.02 | 0.451 | 44.563 | 30.5 | |
| DATG-LModel=Falcon 7B2024.02 | 0.449 | 36.175 | 32.7 | |
| DATG-LModel=Phi-2 2.7B2024.02 | 0.434 | 26.469 | 27.6 | |
| PREADDModel=Phi-2 2.7B2024.02 | 0.433 | 44.667 | 30.1 | |
| FUDGEModel=OPT 6.7B2024.02 | 0.415 | 250.288 | 46 | |
| FUDGEModel=LLaMA-2 13B2024.02 | 0.406 | 256.506 | 42 | |
| FUDGEModel=Falcon 7B2024.02 | 0.403 | 271.234 | 41 | |
| FUDGEModel=Alpaca 7B2024.02 | 0.397 | 271.179 | 42.9 | |
| FUDGEModel=Phi-2 2.7B2024.02 | 0.392 | 267.039 | 42.3 |