Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multiple Choice Question Answering on LongBench v2 (test)
Loading...
44.07
Accuracy (Easy, Short)
Dense
38.7764
40.1507
41.525
42.8993
Jun 3, 2025
Accuracy (Easy, Short)
Accuracy (Easy, Medium)
Accuracy (Easy, Long)
Accuracy (Hard, Short)
Accuracy (Hard, Medium)
Accuracy (Hard, Long)
Accuracy (Total)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy (Easy, Short)
Accuracy (Easy, Medium)
Accuracy (Easy, Long)
Accuracy (Hard, Short)
Accuracy (Hard, Medium)
Accuracy (Hard, Long)
Accuracy (Total)
Dense
Backbone=Llama3.1, Att...
2025.06
44.07
28.41
31.11
32.23
25.98
25.4
30.42
top-k
Backbone=Llama3.1, spa...
2025.06
40.68
25
33.33
29.75
25.2
23.81
28.63
HATA
Backbone=Llama3.1, spa...
2025.06
38.98
27.27
35.56
29.75
26.77
25.4
29.62
Feedback
Search any
task
Search any
task