Locomo
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
LoCoMo
48F1 Score (%)
9
LoCoMo
41.28F1 Score
8
locomo static
0Mean Retrieval Latency (ms)
8
LoCoMo-Plus
74.81Accuracy
8
LoCoMo
80.26Accuracy
8
LoCoMo categories 1-4
91.6Overall Score
8
LoCoMo
69.39Recall@3
8
LoCoMo
29.1BLEU-4
8
LoCoMo
76.7R@5
8
LoCoMo
54.634o-J
8
LoCoMo Open Domain
28.64F1 Score
8
LoCoMo
44.5L Metric
8
LoCoMo
77.4LLM Score
8
LoCoMo (test)
54.17F1 Score
8
LOCOMO full 10-conversation benchmark
79.8LLM-as-Judge Accuracy
7
LOCOMO
59.59SH F1 Score
7
LoCoMo
42.77F1
7
LoCoMo
0.55CSM
7
LoCoMo excluding adversarial (test)
55.5Overall F1
7
LOCOMO (main)
85.37Single Hop Score
7
LoCoMo (test)
80.6LLM Score
7
LoCoMo (test)
86.9LLM Score
7
LoCoMo (test)
59LLM Score
7
LoCoMo (test)
0.742LLM Score
7
LoCoMo (test)
72.3LLM Score
7