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

SIREN

Benchmarks

Task NameDataset NameSOTA ResultTrend
Tool-use agent security evaluationSIREN
Explicit Directive (UA)23.56
16
Fault ClassificationSIREN
IIEE Accuracy (44.1k)100
15
Anomaly DetectionSIREN DCASE Tasks 2020-2025
Performance 2020 (16k)74.26
15
Audio reconstructionSIREN audio segments
Bach MSE (x1e-3)0
5
Showing 4 of 4 rows