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

Multi-document Question Answering on MultidocQA

67.78HotpotQA Accuracy

MemSearch-o1

46.449651.987357.52563.0627Apr 19, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
67.7868.3252.01
2026.04
59.7165.9544.11
2026.04
57.4162.8644.78
2026.04
55.5958.6341.29
2026.04
54.6456.1444.31
2026.04
54.447.2327.54
2026.04
54.2460.0839.95
2026.04
53.9843.7626.95
2026.04
53.7861.1637.61
2026.04
52.1135.7437.99
2026.04
49.5741.3825.67
2026.04
48.4636.7533.32
2026.04
47.5146.6538.05
2026.04
47.2744.8928.26