Natural Language Generation on ToDs benchmark GPT-2 backbone (test)
3.42EERMULTI
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MULTIMemory size (Mem)=Full data, Evaluation type=Modularized2024.05 | 3.42 | 26.15 | |
| HESITMemory size (Mem)=t|M|, Evaluation type=Modularized2024.05 | 5.11 | 21.48 | |
| GSSMemory size (Mem)=t|M|, Evaluation type=Modularized2024.05 | 5.17 | 21.62 | |
| UNIFORMMemory size (Mem)=t|M|, Evaluation type=Modularized2024.05 | 5.42 | 21.17 | |
| RANDOMMemory size (Mem)=t|M|, Evaluation type=Modularized2024.05 | 5.82 | 21.33 | |
| ARPERMemory size (Mem)=t|M|, Evaluation type=Modularized2024.05 | 6.08 | 21.12 | |
| VANILLAMemory size (Mem)=0, Evaluation type=Modularized2024.05 | 18.09 | 10.42 | |
| EWCMemory size (Mem)=0, Evaluation type=Modularized2024.05 | 18.12 | 11.7 | |
| L2Memory size (Mem)=0, Evaluation type=Modularized2024.05 | 18.16 | 11.13 | |
| LAMOLMemory size (Mem)=0, Evaluation type=Modularized2024.05 | 35.83 | 4.43 | |
| A-GEMMemory size (Mem)=t|M|, Evaluation type=Modularized2024.05 | 36.22 | 6.4 |