Question Answering on TriviaQA (F1 score)
39.37F1 ScoreGPT-4o (2024-11-20)
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4o (2024-11-20)Version=2024-11-202025.08 | 39.37 | |
| Llama + LoRRBackbone=Llama3.2-3B, Optimization=LoRR2025.08 | 16.56 | |
| Llama + DPOBackbone=Llama3.2-3B, Optimization=DPO2025.08 | 16.23 | |
| Qwen (Base)Backbone=Qwen2.5-3B, Mode=Base2025.08 | 14.28 | |
| Qwen + LoRRBackbone=Qwen2.5-3B, Optimization=LoRR2025.08 | 14.08 | |
| Qwen + DPOBackbone=Qwen2.5-3B, Optimization=DPO2025.08 | 13.67 | |
| Llama (Base)Backbone=Llama3.2-3B, Mode=Base2025.08 | 12.79 |