Question Answering on OBQA (Client Model Specific Accuracy)
17.8Accuracy (GPT-2-Small)FedCoLLM
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| FedCoLLM2024.11 | 17.8 | 29.2 | 28.8 | |
| Standalone2024.11 | 17.3 | 26.7 | 27.8 | |
| Zero-Shot2024.11 | 16.4 | 23.4 | 23.2 | |
| FedAvg2024.11 | 15.8 | 27.2 | 26 |
Loading the SOTA2 catalog…