Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Question Answering on Wikipedia events 2024-2025 (test)
Loading...
3.78
Standing Q_t^bg
No Wiki
3.3432
3.4566
3.57
3.6834
Jun 3, 2026
Standing Q_t^bg
Event Q_t^vol
Overall Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Standing Q_t^bg
Event Q_t^vol
Overall Score
No Wiki
mode=direct LLM, N=185
2026.06
3.78
3.81
3.8
Online WiCER
signal=AER, N=119
2026.06
3.67
4.62
4.57
FIFO
mode=recency only, N=232
2026.06
3.5
4.77
4.74
Oracle
mode=GT labels, N=232
2026.06
3.5
4.77
4.74
Static Wiki
mode=baseline, N=185
2026.06
3.36
4.64
4.14
Feedback
Search any
task
Search any
task