List your company

Uncertainty Quantification on TQA 16 additional samples (test)

94.1AUROCClustered Self-Assessment
72.88478.39283.989.408Jun 2, 2026

Evaluation results

MethodLinks
Clustered Self-AssessmentBackbone=Qwen2.5-32B2026.0694.1
Clustered Self-AssessmentBackbone=Qwen2.5-14B2026.0693.9
Clustered Self-AssessmentBackbone=Qwen3-14B2026.0693.8
Clustered Self-AssessmentBackbone=Qwen3-30B-A3B2026.0693.5
Clustered Self-AssessmentBackbone=Qwen2.5-7B2026.0693.1
Clustered Self-AssessmentBackbone=Gemma-3-12B2026.0693.1
Clustered Self-AssessmentBackbone=Gemma-3-27B2026.0692.6
DegBackbone=Gemma-3-12B2026.0690.6
sentenceSARBackbone=Qwen3-14B2026.0689.9
LN EntropyBackbone=Qwen3-14B2026.0689.8
SARBackbone=Qwen3-14B2026.0689.8
ProbabilityBackbone=Qwen3-14B2026.0689.7
tokenSARBackbone=Gemma-3-12B2026.0689.7
PerplexityBackbone=Qwen3-14B2026.0689.5
PerplexityBackbone=Gemma-3-12B2026.0689.5
Predictive EntropyBackbone=Qwen3-14B2026.0689.5
PerplexityBackbone=Gemma-3-27B2026.0689.4
DegBackbone=Gemma-3-27B2026.0689.3
tokenSARBackbone=Gemma-3-27B2026.0689.1
ProbabilityBackbone=Qwen2.5-7B2026.0689
P(True)Backbone=Qwen2.5-32B2026.0689
sentenceSARBackbone=Qwen2.5-7B2026.0688.9
PerplexityBackbone=Qwen2.5-7B2026.0688.8
LN EntropyBackbone=Qwen2.5-7B2026.0688.7
EigVBackbone=Qwen3-14B2026.0688.7
Predictive EntropyBackbone=Qwen2.5-7B2026.0688.6
P(True)Backbone=Qwen3-14B2026.0688.6
SARBackbone=Qwen2.5-7B2026.0688.6
sentenceSARBackbone=Qwen2.5-32B2026.0688.5
SARBackbone=Qwen2.5-32B2026.0688.4
EigVBackbone=Gemma-3-12B2026.0688.4
ProbabilityBackbone=Qwen2.5-32B2026.0688.3
ProbabilityBackbone=Gemma-3-27B2026.0688.3
sentenceSARBackbone=Qwen2.5-14B2026.0688.3
SARBackbone=Qwen2.5-14B2026.0688.3
SARBackbone=Qwen3-30B-A3B2026.0688.3
ProbabilityBackbone=Qwen2.5-14B2026.0688.2
LN EntropyBackbone=Qwen2.5-32B2026.0688.2
EigVBackbone=Gemma-3-27B2026.0688.2
Predictive EntropyBackbone=Qwen2.5-32B2026.0688.1
PerplexityBackbone=Qwen2.5-14B2026.0688
LN EntropyBackbone=Qwen2.5-14B2026.0688
PerplexityBackbone=Qwen2.5-32B2026.0687.9
Predictive EntropyBackbone=Qwen2.5-14B2026.0687.9
EigVBackbone=Qwen2.5-7B2026.0687.9
ProbabilityBackbone=Gemma-3-12B2026.0687.8
LN EntropyBackbone=Qwen3-30B-A3B2026.0687.7
EigVBackbone=Qwen2.5-32B2026.0687.7
PerplexityBackbone=Qwen3-30B-A3B2026.0687.6
P(True)Backbone=Qwen3-30B-A3B2026.0687.6
ProbabilityBackbone=Qwen3-30B-A3B2026.0687.5
sentenceSARBackbone=Qwen3-30B-A3B2026.0687.5
Semantic EntropyBackbone=Qwen3-14B2026.0687.5
EccBackbone=Gemma-3-12B2026.0687.5
SARBackbone=Gemma-3-12B2026.0687.4
Eigen ScoreBackbone=Qwen3-14B2026.0687.1
Semantic EntropyBackbone=Qwen2.5-32B2026.0687.1
EccBackbone=Gemma-3-27B2026.0687.1
DegBackbone=Qwen3-14B2026.0687.1
Predictive EntropyBackbone=Qwen3-30B-A3B2026.0687
EigVBackbone=Qwen3-30B-A3B2026.0687
SARBackbone=Gemma-3-27B2026.0686.9
P(True)Backbone=Qwen2.5-14B2026.0686.8
tokenSARBackbone=Qwen3-14B2026.0686.8
Semantic EntropyBackbone=Qwen2.5-7B2026.0686.8
EigVBackbone=Qwen2.5-14B2026.0686.7
Eigen ScoreBackbone=Qwen2.5-7B2026.0686.5
DegBackbone=Qwen2.5-7B2026.0686.2
Semantic EntropyBackbone=Gemma-3-12B2026.0686
Semantic EntropyBackbone=Qwen3-30B-A3B2026.0685.9
DegBackbone=Qwen2.5-32B2026.0685.9
tokenSARBackbone=Qwen2.5-7B2026.0685.7
LN EntropyBackbone=Gemma-3-12B2026.0685.6
Eigen ScoreBackbone=Qwen2.5-32B2026.0685.6
sentenceSARBackbone=Gemma-3-12B2026.0685.6
Semantic EntropyBackbone=Qwen2.5-14B2026.0685.5
tokenSARBackbone=Qwen2.5-14B2026.0685.3
sentenceSARBackbone=Gemma-3-27B2026.0685.3
tokenSARBackbone=Qwen3-30B-A3B2026.0685.1
Eigen ScoreBackbone=Qwen2.5-14B2026.0685
EccBackbone=Qwen3-14B2026.0685
DegBackbone=Qwen3-30B-A3B2026.0684.9
Predictive EntropyBackbone=Gemma-3-12B2026.0684.8
Eigen ScoreBackbone=Qwen3-30B-A3B2026.0684.8
DegBackbone=Qwen2.5-14B2026.0684.8
LN EntropyBackbone=Gemma-3-27B2026.0684.6
tokenSARBackbone=Qwen2.5-32B2026.0684.6
Predictive EntropyBackbone=Gemma-3-27B2026.0684.5
Semantic EntropyBackbone=Gemma-3-27B2026.0684.5
EccBackbone=Qwen2.5-7B2026.0684.1
EccBackbone=Qwen2.5-32B2026.0683.5
P(True)Backbone=Qwen2.5-7B2026.0683.2
Eigen ScoreBackbone=Gemma-3-12B2026.0682.6
EccBackbone=Qwen2.5-14B2026.0682.5
Eigen ScoreBackbone=Gemma-3-27B2026.0682.4
EccBackbone=Qwen3-30B-A3B2026.0682.4
P(True)Backbone=Gemma-3-27B2026.0678.9
P(True)Backbone=Gemma-3-12B2026.0673.7