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

Long-context Question Answering on NarrativeQA (SubEM, F1)

22SubEM

LongMab

912.37515.7519.125Aug 19, 2025
Updated 9d ago

Evaluation Results

MethodLinks
2025.08
2226.83
2025.08
2120.33
2025.08
20.522.67
2025.08
20.523.41
2025.08
2024.37
2025.08
2028.61
2025.08
19.522.54
2025.08
1926.26
2025.08
1921.62
2025.08
1921.17
2025.08
18.527.34
2025.08
18.526.38
2025.08
18.528.05
2025.08
1825.15
2025.08
1826.61
2025.08
1816.27
2025.08
1825.14
2025.08
17.521.52
2025.08
17.519.72
2025.08
1725.86
2025.08
1723.13
2025.08
1724.48
2025.08
16.526.88
2025.08
16.525.92
2025.08
1626.45
2025.08
15.524.98
2025.08
15.521.98
2025.08
15.522.17
2025.08
15.517.67
2025.08
15.518.28
2025.08
1522.31
2025.08
1522.89
2025.08
1518.29
2025.08
1217.03
2025.08
11.518.31
2025.08
9.523.26