Multi-task Generalization on Other tasks (test)
60.28ScoreLM-Cocktail10
Evaluation Results
| Method | Links | |
|---|---|---|
| LM-Cocktail10Fine-tune on=NQ2023.11 | 60.28 | |
| LM-Cocktail10Fine-tune on=Common Gen2023.11 | 58.57 | |
| LM-Cocktail10Fine-tune on=Winogrande2023.11 | 58.52 | |
| LM-Cocktail10Fine-tune on=Hellaswag2023.11 | 57.87 | |
| LM-Cocktail10Fine-tune on=MNLI2023.11 | 56.31 | |
| LM-Cocktail2Fine-tune on=NQ2023.11 | 54.58 | |
| LM-Cocktail10Fine-tune on=SQUAD2023.11 | 54.09 | |
| LM-Cocktail2Fine-tune on=MNLI2023.11 | 53.53 | |
| Fine-tunedFine-tune on=NQ2023.11 | 52.19 | |
| LM-Cocktail2Fine-tune on=SQUAD2023.11 | 51.64 | |
| LM-Cocktail2Fine-tune on=Common Gen2023.11 | 50.88 | |
| LM-Cocktail2Fine-tune on=Winogrande2023.11 | 50.52 | |
| Fine-tunedFine-tune on=SQUAD2023.11 | 49.48 | |
| LM-Cocktail2Fine-tune on=Hellaswag2023.11 | 48.95 | |
| LM-Cocktail10Fine-tune on=AG News2023.11 | 48.32 | |
| LlamaExcluded task=NQ2023.11 | 48.21 | |
| LlamaExcluded task=SQUAD2023.11 | 48.21 | |
| LM-Cocktail2Fine-tune on=AG News2023.11 | 47.73 | |
| Fine-tunedFine-tune on=MNLI2023.11 | 47.49 | |
| LlamaExcluded task=Common Gen2023.11 | 47.48 | |
| Fine-tunedFine-tune on=Winogrande2023.11 | 47.33 | |
| LlamaExcluded task=MRPC2023.11 | 47.11 | |
| LlamaExcluded task=MNLI2023.11 | 47.1 | |
| Fine-tunedFine-tune on=Common Gen2023.11 | 46.9 | |
| LlamaExcluded task=AG News2023.11 | 46.8 | |
| LlamaExcluded task=Winogrande2023.11 | 46.11 | |
| Fine-tunedFine-tune on=Hellaswag2023.11 | 46.1 | |
| LlamaExcluded task=SST22023.11 | 46.02 | |
| LlamaExcluded task=Hellaswag2023.11 | 45.74 | |
| LM-Cocktail10Fine-tune on=SST22023.11 | 45.03 | |
| LM-Cocktail10Fine-tune on=MRPC2023.11 | 42.77 | |
| LM-Cocktail2Fine-tune on=SST22023.11 | 41.63 | |
| LM-Cocktail2Fine-tune on=MRPC2023.11 | 39.56 | |
| Fine-tunedFine-tune on=SST22023.11 | 38.94 | |
| Fine-tunedFine-tune on=AG News2023.11 | 38.58 | |
| Fine-tunedFine-tune on=MRPC2023.11 | 36.45 |