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

Long-context language modeling on RULER

0.9142RULER Score

DASH-3

0.6356880.7079940.78030.852606Feb 6, 2025Apr 25, 2025Jul 12, 2025Sep 28, 2025Dec 15, 2025Mar 3, 2026May 20, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.05
0.9142
2026.05
0.9114
2025.12
0.911
2025.12
0.9061
2026.05
0.9061
2025.12
0.9051
2026.05
0.9051
2025.12
0.8949
2025.12
0.8869
2026.05
0.8869
2026.05
0.8763
2025.12
0.8733
2025.12
0.8717
2026.05
0.8717
2025.12
0.8704
2026.05
0.8688
2025.12
0.8631
2026.05
0.8631
2025.12
0.8619
2025.12
0.8584
2025.12
0.8541
2025.12
0.8533
2025.12
0.8444
2026.05
0.8444
2025.12
0.8423
2025.12
0.8394
2025.12
0.8347
2025.12
0.8259
2026.05
0.8259
2025.12
0.8257
2026.05
0.8257
2025.12
0.8231
2025.12
0.8226
2026.05
0.8226
2025.12
0.8209
2026.05
0.8209
2025.12
0.8205
2026.05
0.8205
2025.12
0.819
2026.05
0.819
2025.12
0.8158
2025.12
0.8126
2025.12
0.8118
2025.12
0.8111
2025.12
0.8031
2026.05
0.8031
2025.12
0.7983
2026.05
0.7983
2025.12
0.7873
2026.05
0.7873
2025.12
0.7728
2025.12
0.7662
2025.12
0.7552
2025.12
0.7539
2025.12
0.7539
2026.05
0.7539
2025.12
0.7538
2025.12
0.7538
2025.12
0.7516
2025.12
0.7453
2025.12
0.7409
2026.05
0.7409
2026.05
0.7358
2025.12
0.7357
2025.12
0.7322
2025.12
0.7318
2026.05
0.7299
2025.12
0.7262
2025.12
0.7227
2025.12
0.7175
2026.05
0.7175
2025.12
0.716
2026.05
0.716
2025.12
0.7152
2025.12
0.7075
2026.05
0.7075
2025.12
0.7065
2025.12
0.6997
2026.05
0.6997
2025.12
0.6991
2025.12
0.6953
2025.02
0.6917
2025.12
0.6907
2025.12
0.6904
2026.05
0.6904
2025.12
0.6786
2025.12
0.6752
2025.12
0.6749
2025.12
0.6743
2025.12
0.6707
2025.12
0.6657
2025.12
0.6626
2025.12
0.6617
2026.05
0.6617
2025.12
0.659
2025.12
0.6569
2025.12
0.6544
2025.12
0.6479
2025.12
0.6469
2025.12
0.6464
Showing 100 of 204 rows