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

Question Answering on NarrativeQA LongBench (32K Context)

17.2F1 Score

Full cache

0.1444.572913.428May 18, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.05
17.2100-
2026.05
10.862.8-
2026.05
10.359.90.873
2026.05
8.6112.1-
2026.05
8.6111.90.893
2026.05
8.6111.90.893
2026.05
8.3107.20.99
2026.05
7.7100-
2026.05
1.722.40.001
2026.05
1.721.90.001
2026.05
1.317.50.001
2026.05
0.84.60.001