List your company

Hallucination Detection on TriviaQA (40% test)

91AUROCACSE
53.5663.287382.72May 5, 2026

Evaluation results

MethodLinks
ACSELLM Architecture=Qwen-7B2026.059125198990
ACSELLM Architecture=Mistral-7B2026.058928248887
ACSELLM Architecture=LLaMA-2-7B2026.058635298384
ACSELLM Architecture=Falcon-7B2026.058438328182
CAPLLM Architecture=Qwen-7B2026.058443368183
DDCRP-CPLLM Architecture=Qwen-7B2026.058247407981
CAPLLM Architecture=Mistral-7B2026.058048417976
CAPLLM Architecture=LLaMA-2-7B2026.057948417676
SULLM Architecture=Qwen-7B2026.057950437778
DDCRP-CPLLM Architecture=Mistral-7B2026.057752467674
DDCRP-CPLLM Architecture=LLaMA-2-7B2026.057752457474
CAPLLM Architecture=Falcon-7B2026.057654467374
SULLM Architecture=LLaMA-2-7B2026.057556497272
DDCRP-CPLLM Architecture=Falcon-7B2026.057457497172
EigVLLM Architecture=Qwen-7B2026.057461547172
SULLM Architecture=Mistral-7B2026.057355497372
SULLM Architecture=Falcon-7B2026.057260537068
EigVLLM Architecture=Mistral-7B2026.057165586969
P(True)LLM Architecture=Mistral-7B2026.057072656865
P(True)LLM Architecture=Qwen-7B2026.057072656768
EigVLLM Architecture=LLaMA-2-7B2026.056966596667
EigVLLM Architecture=Falcon-7B2026.056772646264
TELLM Architecture=Mistral-7B2026.056578726361
P(True)LLM Architecture=LLaMA-2-7B2026.056577696161
TELLM Architecture=Qwen-7B2026.056480726162
P(True)LLM Architecture=Falcon-7B2026.056181745859
TELLM Architecture=LLaMA-2-7B2026.055985785655
TELLM Architecture=Falcon-7B2026.055589825352