Natural Language Inference on RTE SuperGLUE (test)
66.13AccuracyFedAvg
Evaluation results
| Method | Links | ||
|---|---|---|---|
| FedAvgModel=OPT-1.3B2024.05 | 66.13 | — | |
| FedAvgModel=OPT-125M2024.05 | 63.96 | — | |
| DeComFLModel=OPT-1.3B, P=102024.05 | 60.89 | — | |
| FedZOModel=OPT-1.3B, P=102024.05 | 59.21 | — | |
| MeZOModel=OPT-1.3B, Setting=single agent2024.05 | 57.57 | — | |
| DeComFLModel=OPT-125M, P=102024.05 | 57.05 | — | |
| FedZOModel=OPT-125M, P=102024.05 | 54.16 | — | |
| MeZOModel=OPT-125M, Setting=single agent2024.05 | 52.91 | — |