List your company

Faithfulness Evaluation on WorldCapital Conflicting short-context

127.9AOPCcompATTN
95.14103.645112.15120.655Oct 3, 2025

Evaluation results

MethodLinks
ATTNModel=Qwen2.5-7B2025.10127.9153.76
FAModel=Qwen2.5-7B2025.10122.7150.61
MechLightModel=Qwen2.5-7B2025.10119.3151.61
IGModel=Qwen2.5-7B2025.10118149.33
IGModel=Pythia-6.9B2025.10114.5147.82
FAModel=Pythia-6.9B2025.10113.6147.09
MechLightModel=Pythia-6.9B2025.10104.2151.9
ATTNModel=Pythia-6.9B2025.1096.4151.13