Faithfulness Evaluation on AVeriTeC
0.3FIDGemma-2
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gemma-2Parameters=2B2026.03 | 0.3 | 0.13 | 0.17 | |
| Llama-3.2Parameters=3B2026.03 | 0.47 | 0.08 | 0.39 | |
| Llama-3.1Parameters=8B2026.03 | 0.81 | 0.26 | 0.55 | |
| Falcon-3Parameters=3B2026.03 | 0.82 | 0.37 | 0.45 | |
| Qwen-3Parameters=1.7B2026.03 | 0.83 | 0.21 | 0.62 | |
| Falcon-3Parameters=7B2026.03 | 0.86 | 0.46 | 0.4 | |
| Qwen-3Parameters=4B2026.03 | 0.92 | 0.34 | 0.58 | |
| Qwen-3Parameters=8B2026.03 | 0.93 | 0.29 | 0.64 |