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

Multi-document Question Answering on LongBench Multi-Doc QA v2

60.3Accuracy

DELM+RLM

20.05230.50140.9551.399Jun 9, 2026Jun 12, 2026Jun 16, 2026Jun 20, 2026Jun 23, 2026Jun 27, 2026Jul 1, 2026
Updated 23d ago

Evaluation Results

MethodLinks
2026.06
60.30.24-
2026.06
57.90.3-
2026.06
55.80.29-
2026.07
47.2-104.2
2026.07
40-235.4
2026.07
35-58.8
2026.07
21.6-104.8