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

Multi-page Question Answering on LongBench multi-page

48.9HQA F1

AGAR

23.9430.4236.943.38Jun 11, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
48.943.825.638-
2026.06
48.643.12437.2-
2026.06
48.643.12437.2-
2026.06
48.331.37.121-
2026.06
47.542.923.837.3-
2026.06
46.950.726.825.9-
2026.06
46.750.224.723.6-
2026.06
45.949.923.924.9-
2026.06
45.824.721.919.9-
2026.06
45.738.926.518.6-
2026.06
45.650.324.522.7-
2026.06
44.550.124.824.8-
2026.06
44.550.224.824.2-
2026.06
44.340.822.134.2-
2026.06
43.546.525.523.9-
2026.06
42.136.118.229.9-
2026.06
41.835.221.330.3-
2026.06
41.239.121.731.5-
2026.06
40.837.317.633.8-
2026.06
39.837.517.229.5-
2026.06
39.837.516.933.3-
2026.06
37.739.218.530.9-
2026.06
3732.910.122.4-
2026.06
34.136.621.229.3-
2026.06
31.729.517.817.9-
2026.06
3030.213.412.4-
2026.06
24.923.89.519.5-