Question Answering on ARC-E (Client Model Specific Accuracy)
46.8Accuracy (GPT-2-Small)FedCoLLM
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| FedCoLLM2024.11 | 46.8 | 60.01 | 61.1 | |
| Standalone2024.11 | 46.4 | 59.93 | 60.1 | |
| FedAvg2024.11 | 46.3 | 59.97 | 60.3 | |
| Zero-Shot2024.11 | 43.8 | 57 | 53.1 |