List your company

Hallucination Detection on FaithBench

61.11Consistency RateGLM-4-9B-Chat
36.555642.930349.30555.6797Mar 25, 2026

Evaluation results

MethodLinks
GLM-4-9B-ChatParameters=9B2026.0361.11
MARCH-STEMBackbone=Meta-Llama3.1-8B-Instruct2026.0356.94
o4-mini-medium-2025-04-162026.0355.56
o4-mini-low-2025-04-162026.0355.56
MARCH-GeneralBackbone=Meta-Llama3.1-8B-Instruct2026.0352.78
o4-mini-high-2025-04-162026.0345.83
Qwen2.5-14B-InstructParameters=14B2026.0345.83
Ministral-8B-InstructParameters=8B2026.0343.06
Phi-4-mini-Instruct2026.0338.89
Meta-Llama3.1-8B-InstructParameters=8B2026.0337.5