List your company

Hallucination Detection on GSM8K (Comprehensive Metrics)

80.62AUROCHALLUGUARD
64.791268.900673.0177.1194Jan 26, 2026

Evaluation results

MethodLinks
HALLUGUARDBackbone=Mistral-7B2026.0180.6277.380.2276.8868.21
HALLUGUARDBackbone=LLaMA2-13B2026.0179.0176.7379.174.2567.92
SAPLMABackbone=Mistral-7B2026.0177.17476.269.559.8
HALLUGUARDBackbone=QwQ-32B2026.0175.8174.6882.5578.9170.45
SAPLMABackbone=LLaMA2-13B2026.0175.472.37467.159.1
ITIBackbone=Mistral-7B2026.0174.270.667.160.850.1
SAPLMABackbone=QwQ-32B2026.0173.971.276.570.160.7
LLM-CheckBackbone=Mistral-7B2026.0173.469.864.957.948.3
ITIBackbone=LLaMA2-13B2026.017369.166.158.449.5
HALLUGUARDBackbone=OPT-6.7B2026.0172.5770.3174.9169.1462.1
ITIBackbone=QwQ-32B2026.0172.569.269.462.552.3
LLM-CheckBackbone=LLaMA2-13B2026.0172.168.564.256.648
HALLUGUARDBackbone=GPT22026.0172.0469.8878.3374.1165.42
LLM-CheckBackbone=QwQ-32B2026.0171.267.967.860.350.4
SAPLMABackbone=OPT-6.7B2026.0170.867.271.864.156.3
SAPLMABackbone=GPT22026.0169.266.170.16254.4
ITIBackbone=OPT-6.7B2026.016965.469.261.551.9
LLM-CheckBackbone=OPT-6.7B2026.0168.164.367.559.249.8
ITIBackbone=GPT22026.0166.86364.556.248.7
LLM-CheckBackbone=GPT22026.0165.461.562.454.146.2