List your company

Hallucination Evaluation on MCEval HaluEval 8K (test)

82.2AccuracyAct
50.79258.94667.175.254Dec 24, 2024

Evaluation results

MethodLinks
ActModel=Llama3.2-3B2024.1282.2
ActModel=Qwen2.5-7B2024.1282.1
ActModel=Llama3.1-8B2024.1280.7
Magn-ProbeModel=Llama3.1-8B2024.1279.8
ActModel=Llama2-7B2024.1278.8
Magn-ProbeModel=Llama2-7B2024.1278.3
Magn-ProbeModel=Llama3.2-3B2024.1278
Magn-ProbeModel=Qwen2.5-7B2024.1277.3
Magn-ProbeModel=Llama2-70B2024.1276.3
LM-ProbModel=Llama2-70B2024.1275.4
LM-ProbModel=Llama3.1-8B2024.1267.4
LM-ProbModel=Llama3.2-3B2024.1265
LM-ProbModel=Qwen2.5-7B2024.1263.7
LM-ProbModel=Llama2-7B2024.1252