List your company

Hallucination Detection on FinanceBench

73.4F1 ScoreHalluClean
4.7622.5840.458.22Nov 12, 2025

Evaluation results

MethodLinks
HalluCleanBackbone=GPT-3.5-turbo2025.1173.476.5
DeepSeek-R12025.1147.963
DeepSeek-V32025.1121.155
GPT-4o-mini2025.1119.450
GPT-3.5-turbo2025.111151.5
Llama-3-70B2025.117.450