List your company

Uncertainty Quantification on TQA 4 additional samples (test)

0.937AUROCClustered Self-Assessment
0.7290.7830.8370.891Jun 2, 2026

Evaluation results

MethodLinks
Clustered Self-AssessmentModel=Qwen2.5-32B2026.060.937
Clustered Self-AssessmentModel=Qwen2.5-14B2026.060.936
Clustered Self-AssessmentModel=Qwen3-14B2026.060.932
Clustered Self-AssessmentModel=Qwen3-30B-A3B2026.060.929
Clustered Self-AssessmentModel=Gemma-3-12B2026.060.927
Clustered Self-AssessmentModel=Gemma-3-27B2026.060.922
Clustered Self-AssessmentModel=Qwen2.5-7B2026.060.921
SARModel=Gemma-3-12B2026.060.904
DegModel=Gemma-3-12B2026.060.898
ProbabilityModel=Qwen3-14B2026.060.897
tokenSARModel=Gemma-3-12B2026.060.897
sentenceSARModel=Qwen3-14B2026.060.897
PerplexityModel=Qwen3-14B2026.060.895
PerplexityModel=Gemma-3-12B2026.060.895
LN EntropyModel=Qwen3-14B2026.060.895
PerplexityModel=Gemma-3-27B2026.060.894
Predictive EntropyModel=Qwen3-14B2026.060.894
SARModel=Gemma-3-27B2026.060.892
tokenSARModel=Gemma-3-27B2026.060.891
ProbabilityModel=Qwen2.5-7B2026.060.89
P(True)Model=Qwen2.5-32B2026.060.89
SARModel=Qwen3-14B2026.060.89
EigVModel=Gemma-3-12B2026.060.889
PerplexityModel=Qwen2.5-7B2026.060.888
sentenceSARModel=Qwen2.5-7B2026.060.887
LN EntropyModel=Qwen2.5-7B2026.060.886
P(True)Model=Qwen3-14B2026.060.886
Predictive EntropyModel=Qwen2.5-7B2026.060.885
LN EntropyModel=Gemma-3-12B2026.060.885
sentenceSARModel=Qwen2.5-32B2026.060.884
ProbabilityModel=Qwen2.5-32B2026.060.883
ProbabilityModel=Gemma-3-27B2026.060.883
ProbabilityModel=Qwen2.5-14B2026.060.882
LN EntropyModel=Qwen2.5-32B2026.060.881
sentenceSARModel=Qwen2.5-14B2026.060.881
DegModel=Gemma-3-27B2026.060.881
PerplexityModel=Qwen2.5-14B2026.060.88
Predictive EntropyModel=Qwen2.5-32B2026.060.88
SARModel=Qwen2.5-7B2026.060.88
PerplexityModel=Qwen2.5-32B2026.060.879
LN EntropyModel=Qwen2.5-14B2026.060.879
sentenceSARModel=Gemma-3-12B2026.060.879
ProbabilityModel=Gemma-3-12B2026.060.878
Predictive EntropyModel=Qwen2.5-14B2026.060.878
EigVModel=Gemma-3-27B2026.060.878
SARModel=Qwen2.5-32B2026.060.877
PerplexityModel=Qwen3-30B-A3B2026.060.876
P(True)Model=Qwen3-30B-A3B2026.060.876
sentenceSARModel=Gemma-3-27B2026.060.876
ProbabilityModel=Qwen3-30B-A3B2026.060.875
SARModel=Qwen2.5-14B2026.060.875
LN EntropyModel=Qwen3-30B-A3B2026.060.874
SARModel=Qwen3-30B-A3B2026.060.874
sentenceSARModel=Qwen3-30B-A3B2026.060.873
LN EntropyModel=Gemma-3-27B2026.060.87
Predictive EntropyModel=Qwen3-30B-A3B2026.060.869
Predictive EntropyModel=Gemma-3-12B2026.060.869
P(True)Model=Qwen2.5-14B2026.060.868
tokenSARModel=Qwen3-14B2026.060.868
Predictive EntropyModel=Gemma-3-27B2026.060.865
Semantic EntropyModel=Gemma-3-12B2026.060.859
tokenSARModel=Qwen2.5-7B2026.060.857
Semantic EntropyModel=Gemma-3-27B2026.060.855
tokenSARModel=Qwen2.5-14B2026.060.853
EigVModel=Qwen3-14B2026.060.852
tokenSARModel=Qwen3-30B-A3B2026.060.851
tokenSARModel=Qwen2.5-32B2026.060.846
EigVModel=Qwen2.5-7B2026.060.841
EigVModel=Qwen2.5-32B2026.060.841
Semantic EntropyModel=Qwen3-14B2026.060.839
DegModel=Qwen3-14B2026.060.839
EccModel=Gemma-3-27B2026.060.837
EccModel=Gemma-3-12B2026.060.834
P(True)Model=Qwen2.5-7B2026.060.832
Semantic EntropyModel=Qwen2.5-7B2026.060.829
DegModel=Qwen2.5-7B2026.060.828
DegModel=Qwen2.5-32B2026.060.828
EigVModel=Qwen3-30B-A3B2026.060.827
Semantic EntropyModel=Qwen2.5-32B2026.060.826
Eigen ScoreModel=Qwen3-14B2026.060.82
EigVModel=Qwen2.5-14B2026.060.82
Semantic EntropyModel=Qwen3-30B-A3B2026.060.817
Eigen ScoreModel=Qwen2.5-7B2026.060.816
Eigen ScoreModel=Gemma-3-12B2026.060.814
EccModel=Qwen3-14B2026.060.814
DegModel=Qwen3-30B-A3B2026.060.813
Eigen ScoreModel=Qwen2.5-32B2026.060.81
DegModel=Qwen2.5-14B2026.060.809
Semantic EntropyModel=Qwen2.5-14B2026.060.807
Eigen ScoreModel=Gemma-3-27B2026.060.805
EccModel=Qwen2.5-7B2026.060.805
EccModel=Qwen2.5-32B2026.060.801
Eigen ScoreModel=Qwen3-30B-A3B2026.060.798
Eigen ScoreModel=Qwen2.5-14B2026.060.796
P(True)Model=Gemma-3-27B2026.060.789
EccModel=Qwen2.5-14B2026.060.785
EccModel=Qwen3-30B-A3B2026.060.785
P(True)Model=Gemma-3-12B2026.060.737