List your company

Multi-hop Faithfulness Hallucination Detection on HoVer Refined

82.9Macro F1FaithLens
72.70875.3547880.646Dec 23, 2025

Evaluation results

MethodLinks
FaithLensbackbone=Llama-3.1-8B-Inst2025.1282.9
GPT-4.12025.1282.6
Llama-3.1-405B-Inst2025.1281.6
o32025.1281.1
ClearCheckbackbone=Llama-3.1-8B-Inst2025.1280.3
Claude-3.7-Sonnet2025.1280.2
DeepSeek-V3.2thinking=true2025.1280
o12025.1279.9
o3-mini2025.1278.5
DeepSeek-V3.2thinking=false2025.1276.7
MiniCheckparameters=7B2025.1274.9
GPT-4o2025.1273.6
AlignScoreparameters=355M2025.1273.3
FactCGparameters=435M2025.1273.1