Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Question Answering on LongBench NarrativeQA
Loading...
256
Tokens
InfLLM
23.04
83.52
144
204.48
Mar 19, 2026
Tokens
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Tokens
Accuracy
InfLLM
Settings=K=16
2026.03
256
15.57
UT-ACA
Settings=Kmax=32
2026.03
252
17.81
InfLLM
Settings=K=8
2026.03
128
15.75
UT-ACA
Settings=Kmax=16
2026.03
109
16.92
InfLLM
Settings=K=4
2026.03
64
14.95
UT-ACA
Settings=Kmax=8
2026.03
59
16.11
UT-ACA
Settings=Kmax=4
2026.03
39
15.01
InfLLM
Settings=K=2
2026.03
32
13.49
Feedback
Search any
task
Search any
task
