Locomo
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
LoCoMo Single-hop
34.1Mean Token F1
23
LoCoMo
84Accuracy
22
LoCoMo
97.5Accuracy (Single-Hop)
22
LoCoMo full
61.56Single-Hop F1
21
LoCoMo tau = 96k (test)
495.2E2E Wall Time (s)
20
LoCoMo
72.3Org. Score
20
LOCOMO (Evaluation)
48.01Single-Hop Retrieval F1
20
LoCoMo Multi-Hop
38.72F1 Score
20
LoCoMo 32K
46.39F1 Score
20
LoCoMo
0.257Multi-Hop RGE-L
20
LoCoMo Overall All Categories 1.0
4.63EM Rank
20
LoCoMo Single-Hop 1.0
16.77EM
20
LoCoMo Open-Domain 1.0
7.29EM
20
LoCoMo Temporal 1.0
1,121EM
20
LoCoMo Multi-Hop 1.0
426EM
20
LoCoMo Overall
1Avg Rank (F1)
20
LoCoMo Single-Hop
42.39F1 Score
20
LoCoMo Open-Domain
20.73F1
20
LoCoMo Temporal
50.83F1 Score
20
LoCoMo Multi-Hop
26F1 Score
20
LoCoMo
43.05F1 Score
20
LoCoMo Single-Hop
54.57F1 Score
19
LoCoMo Temporal
59.1F1 Score
19
LoCoMo
31.3Precision
18
LoCoMo (test)
33.09MultiHop BLEU
18