Faithfulness Evaluation on WorldCapital Double-Conflicting short-context
187.7AOPCcompFA
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| FAModel=Pythia-6.9B2025.10 | 187.7 | 267.13 | |
| IGModel=Pythia-6.9B2025.10 | 186.7 | 267.9 | |
| ATTNModel=Qwen2.5-7B2025.10 | 184.4 | 245.35 | |
| FAModel=Qwen2.5-7B2025.10 | 182.7 | 250.98 | |
| IGModel=Qwen2.5-7B2025.10 | 180.7 | 255.75 | |
| MechLightModel=Qwen2.5-7B2025.10 | 177.3 | 244.78 | |
| MechLightModel=Pythia-6.9B2025.10 | 174.8 | 265.56 | |
| ATTNModel=Pythia-6.9B2025.10 | 155.4 | 267.18 |