Long-term Memory Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
98.7KU Accuracy
16
Jun 16, 2026
77.8QS
15
May 6, 2026
94.48Accuracy
10
May 7, 2026
59.59SH F1 Score
7
May 29, 2026
78.7SSU F1
7
Jun 23, 2026
37.41Multi-hop F1
7
May 18, 2026
77.53Average Judge Score
4
Jun 4, 2026
87.16Average Judge Score
2
Jun 4, 2026
70.72Average Judge Score
2
Jun 4, 2026