List your company

Hallucination Detection on CoQA (AUCs, AUCr, PCC)

77.5AUCsLSC
24.4638.235265.77Jan 7, 2026

Evaluation results

MethodLinks
LSCModel=Qwen-32B2026.0177.57541.2
LN-EntropyModel=Qwen-32B2026.0175.37330.1
AGSERModel=Qwen-32B2026.0174.671.230.1
Lexical SimilarityModel=Qwen-32B2026.017371.934
EigenScoreModel=Qwen-0.5B2026.017271.838.4
Lexical SimilarityModel=Qwen-7B2026.0171.671.834.6
EigenScoreModel=Qwen-7B2026.0171.57138.7
LSCModel=Qwen-7B2026.0171.570.934.7
EigenScoreModel=LLaMA-13B2026.0171.473.440.5
EigenScoreModel=LLaMA-7B2026.0171.372.636.3
EigenScoreModel=Qwen-32B2026.0171.270.741.1
LSCModel=LLaMA-13B2026.0169.573.937.8
LN-EntropyModel=Qwen-7B2026.0169.469.227.9
LSCModel=Qwen-3B2026.0169.369.331.4
AGSERModel=Qwen-7B2026.0169.169.127.4
Lexical SimilarityModel=LLaMA-13B2026.0168.772.936.1
EigenScoreModel=Qwen-3B2026.0168.46835
Lexical SimilarityModel=LLaMA-7B2026.0168.271.729.9
LSCModel=LLaMA-7B2026.0167.772.131.3
Lexical SimilarityModel=Qwen-0.5B2026.0167.26931.7
Lexical SimilarityModel=Qwen-3B2026.0167.267.929.1
AGSERModel=Qwen-3B2026.0167.266.725.7
AGSERModel=Qwen-0.5B2026.016766.125.7
PerplexityModel=Qwen-32B2026.0166.668.319.9
LSCModel=Qwen-0.5B2026.0166.469.132.8
AGSERModel=LLaMA-13B2026.0165.76625.5
LN-EntropyModel=Qwen-3B2026.0165.165.719.6
AGSERModel=LLaMA-7B2026.0162.26320
PerplexityModel=Qwen-7B2026.0160.261.911.8
LN-EntropyModel=LLaMA-13B2026.015863.816.5
LN-EntropyModel=Qwen-0.5B2026.015658.69.7
LN-EntropyModel=LLaMA-7B2026.0155.362.18.4
PerplexityModel=Qwen-3B2026.015558.58.3
PerplexityModel=LLaMA-13B2026.0151.659.14.7
PerplexityModel=Qwen-0.5B2026.0148.954.94.2
PerplexityModel=LLaMA-7B2026.0147.956.80.1
EnergyModel=LLaMA-7B2026.0144.352.3-7.1
EnergyModel=LLaMA-13B2026.0143.750.5-9.7
EnergyModel=Qwen-0.5B2026.0143.147.2-8.2
EnergyModel=Qwen-3B2026.0141.144.2-10.1
EnergyModel=Qwen-7B2026.0136.839.8-17
EnergyModel=Qwen-32B2026.0126.536.7-20.3