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

Hallucination Detection on SQuAD (AUROC)

0.89AUROC

Updated 1mo ago

Evaluation Results

Method	Links
ID 2026.05		0.89
FEPoID 2026.05		0.89
ID 2026.05		0.8884
FEPoID 2026.05		0.8884
Val Loss 2026.05		0.887
Val Loss 2026.05		0.8865
Val Loss 2026.05		0.8784
ID 2026.05		0.8769
FEPoID 2026.05		0.8769
Attention probe, soft target 2026.06		0.8754
Curvature 2026.05		0.8704
Val Loss 2026.05		0.8679
ID 2026.05		0.8679
FEPoID 2026.05		0.8679
Curvature 2026.05		0.8625
Curvature 2026.05		0.8613
Attention probe, soft target 2026.06		0.8566
Curvature 2026.05		0.8553
RankME 2026.05		0.854
Attention probe 2026.06		0.8519
Attention probe 2026.06		0.8479
Lexical Similarity 2026.05		0.8433
Attention probe 2026.06		0.8339
Attention probe, soft target 2026.06		0.8333
Attention probe, soft target 2026.06		0.8213
EigenScore 2026.05		0.8152
Attention probe 2026.06		0.8149
RankME 2026.05		0.8106
RGN 2026.05		0.8104
Attention probe, soft target 2026.06		0.8015
Attention probe, soft target 2026.06		0.8006
Attention probe 2026.06		0.7977
Linear probe 2026.06		0.7967
Entropy 2026.06		0.7954
RGN 2026.05		0.7929
ID 2026.05		0.7892
FEPoID 2026.05		0.7892
Attention probe, soft target 2026.06		0.7852
Pred. Entropy 2026.05		0.785
Val Loss 2026.05		0.7811
Semantic Entropy 2026.05		0.7806
Attention probe, soft target 2026.06		0.78
Attention probe, soft target 2026.06		0.7701
Attention probe 2026.06		0.7695
Linear probe 2026.06		0.7681
Pred. Entropy 2026.05		0.7633
RGN 2026.05		0.7553
RankME 2026.05		0.7548
Attention probe 2026.06		0.7532
EigenScore 2026.05		0.7508
LN-Entropy 2026.05		0.748
Curvature 2026.05		0.7432
LN-Pred. Entropy 2026.05		0.7425
Attention probe 2026.06		0.7393
Linear probe 2026.06		0.7372
EigenScore 2026.05		0.735
RankME 2026.05		0.7341
Attention probe 2026.06		0.7323
Pred. Entropy 2026.05		0.7316
RGN 2026.05		0.7313
RGN 2026.05		0.7305
ID 2026.05		0.7275
FEPoID 2026.05		0.7275
Lexical Similarity 2026.05		0.727
Entropy 2026.06		0.7254
SNR 2026.05		0.723
Lexical Similarity 2026.05		0.7169
Val Loss 2026.05		0.7153
RankME 2026.05		0.7132
SNR 2026.05		0.7112
Linear probe 2026.06		0.7027
Entropy 2026.06		0.6963
Attention probe, soft target 2026.06		0.6924
Linear probe 2026.06		0.6899
SNR 2026.05		0.687
EmbRegress 2026.05		0.678
SNR 2026.05		0.6693
Curvature 2026.05		0.6616
RGN 2026.05		0.6567
LN-Pred. Entropy 2026.05		0.6564
Attention probe 2026.06		0.6539
Linear probe 2026.06		0.6537
FEPoID 2026.05		0.6514
RankME 2026.05		0.6476
Curvature 2026.05		0.6442
Val Loss 2026.05		0.6442
Entropy 2026.06		0.6426
Semantic Entropy 2026.05		0.6409
Linear probe 2026.06		0.6399
LN-Pred. Entropy 2026.05		0.639
SNR 2026.05		0.6387
FEPoID 2026.05		0.6377
Semantic Entropy 2026.05		0.6314
SNR 2026.05		0.6225
Curvature 2026.05		0.6183
Val Loss 2026.05		0.6164
RGN 2026.05		0.613
ID 2026.05		0.613
RankME 2026.05		0.6071
Lexical Similarity 2026.05		0.5988

Showing 100 of 127 rows