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

Long-context Language Modeling on LongBench

46.23LongBench Average Score

AdaKV w/ CriticalKV

16.43424.169531.90539.6405Feb 6, 2025Apr 25, 2025Jul 12, 2025Sep 29, 2025Dec 16, 2025Mar 4, 2026May 22, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2025.02
46.23-------
2025.02
46.1-------
2025.02
45.35-------
2025.02
44.23-------
2025.02
44.03-------
2026.05
19.3-------
2026.05
19.3-------
2026.05
19.25-------
2026.05
19.19-------
2026.05
19.12-------
2026.05
19.02-------
2026.05
17.58-------
2026.01
-42.0486.223.2372.6731.1110059.21
2026.01
-32.5161.5610.4352.6713.086.229.41
2026.01
-40.5384.3423.766826.1910057.14