Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Magpie

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multi-agent Sequential CommunicationMAGPIE
Privacy Score88
10
IdiomsMAGPIE
AUROC98.8
8
LLM RoutingMagpie Out-of-Domain
LPM Score63.53
7
LLM RoutingMagpie Out-of-Domain
AUROC74.08
7
Multi-agent latent communication privacy and utilityMAGPIE Graph
Privacy80
5
Multi-agent latent communication privacy and utilityMAGPIE Hierarchical
Privacy82
5
Magnetic Indoor LocalizationMagPie Loomis building
MAE1.07
4
Magnetic Indoor LocalizationMagPie Talbot building
MAE0.64
4
Magnetic Indoor LocalizationMagPie CSL building
MAE0.21
4
Showing 9 of 9 rows