LongMemEval
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
LongMemEval
95.6Average Score
103
LongMemEval LongConvQA
90.3SH Score
84
LongMemEval 100 question sample
100Evidence-round Recall
48
LongMemEval Small
51.2F1 Score (%)
40
LONGMEMEVAL-S
73.2Accuracy
31
LongMemEval
90.8Score
30
LongMemEval S (test)
94.4KU (Knowledge Update)
30
LongMemEval
99Recall@5
25
LongMemEval
34.72Accuracy
25
LongMemEval
65.92EM
24
LongMemEval-S (N=500)
91Temporal Accuracy
24
LongMemEvalS
95.6Overall Score
23
LongMemEval
76Accuracy
22
LongMemEval 1M
49.58F1 Score
20
LongMemEval 128K
47.26F1 Score
20
LongMemEval
59.9Top-5 Recall
20
LongMemEval
40.53F1 Score
20
LongMemEval S (test)
84.21QA Score (TR Context)
19
LongMemEval-S
100SSU
19
LongMemEval
68.5Precision
18
LongMemEval (test)
85.75Accuracy
18
LongMemEval S
97.14Single-User Score
17
LongMemEval-S
94.68Recall@5
17
LongMemEval-RR
46.33F1 Score
16
LongMemEval-S (500 questions)
98.7KU Accuracy
16