Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Question Answering on RULER QA-16k (test)
Loading...
512
Token Count
InfLLM
29.44
154.72
280
405.28
Mar 19, 2026
Token Count
Mean Concatenated Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Token Count
Mean Concatenated Accuracy
InfLLM
Token Budget=512
2026.03
512
21.7
UT-ACA
Token Budget=512
2026.03
385
21.65
InfLLM
Token Budget=256
2026.03
256
18.75
UT-ACA
Token Budget=256
2026.03
203
17.9
InfLLM
Token Budget=128
2026.03
128
14.05
UT-ACA
Token Budget=128
2026.03
105
14.65
InfLLM
Token Budget=64
2026.03
64
9.25
UT-ACA
Token Budget=64
2026.03
48
10.09
Feedback
Search any
task
Search any
task