A Stack-Propagation Framework with Token-Level Intent Detection for Spoken Language Understanding

Benchmark Model Rank Results
intent-detection-on-atisStack-Propagation (+BERT)#8Accuracy: 97.50
intent-detection-on-snipsStack-Propagation (+BERT)#2Accuracy: 99.0
intent-detection-on-snipsStack-Propagation#5Accuracy: 98.00
slot-filling-on-atisStack-Propagation (+BERT)#4F1: 0.9610
slot-filling-on-snipsStack-Propagation (+BERT)#2F1: 97.00
slot-filling-on-snipsStack-Propagation#5F1: 94.20