Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Hallucination Detection on TruthfulQA

0.9417AUC (ROC)

ARS (CCS)

0.3679320.5168910.665850.814809Dec 1, 2025Dec 23, 2025Jan 14, 2026Feb 5, 2026Feb 27, 2026Mar 21, 2026Apr 12, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.01
0.9417
2026.01
0.9375
2026.01
0.9293
2025.12
0.9268
2026.01
0.913
2025.12
0.9082
2025.12
0.8993
2025.12
0.8991
2026.01
0.8664
2026.03
0.839
2026.01
0.8366
2026.04
0.821
2026.01
0.8195
2026.03
0.818
2026.04
0.811
2026.01
0.8089
2026.04
0.806
2026.04
0.804
2026.04
0.798
2026.04
0.797
2026.03
0.795
2026.01
0.7946
2025.12
0.7864
2026.04
0.785
2026.04
0.779
2026.04
0.778
2026.04
0.775
2026.04
0.774
2026.03
0.773
2026.04
0.773
2026.04
0.773
2026.03
0.772
2026.04
0.771
2026.01
0.7708
2026.04
0.77
2026.01
0.7698
2026.04
0.767
2026.03
0.765
2026.04
0.765
2026.04
0.765
2026.04
0.761
2026.04
0.757
2026.04
0.757
2026.04
0.755
2026.03
0.753
2026.03
0.75
2026.04
0.748
2026.03
0.741
2026.04
0.741
2026.03
0.732
2025.12
0.7317
2026.01
0.7308
2026.01
0.7258
2026.03
0.724
2025.12
0.72
2026.01
0.7186
2026.03
0.713
2026.04
0.704
2026.01
0.7001
2026.01
0.6949
2026.04
0.693
2025.12
0.6795
2026.04
0.678
2026.01
0.6757
2026.03
0.673
2026.04
0.667
2026.03
0.663
2026.03
0.661
2026.03
0.661
2026.01
0.656
2026.03
0.655
2026.03
0.642
2025.12
0.6391
2026.03
0.635
2026.03
0.633
2026.03
0.632
2026.01
0.6244
2026.03
0.622
2026.03
0.621
2025.12
0.62
2026.03
0.603
2026.01
0.6024
2026.03
0.598
2026.01
0.5972
2026.03
0.597
2026.01
0.5881
2026.01
0.5864
2026.01
0.5664
2026.01
0.5662
2026.01
0.5614
2026.01
0.5595
2026.01
0.5547
2026.01
0.5215
2026.03
0.509
2025.12
0.5
2026.01
0.5
2026.01
0.4537
2025.12
0.45
2025.12
0.41
2025.12
0.39
Showing 100 of 102 rows