Abstract | ||
---|---|---|
This paper presents a hybrid dialog state tracker enhanced by trainable Spoken Language Understanding (SLU) for slot-filling dialog systems. Our architecture is inspired by previously proposed neural-network-based belief-tracking systems. In addition, we extended some parts of our modular architecture with differentiable rules to allow end-to-end training. We hypothesize that these rules allow our tracker to generalize better than pure machine-learning based systems. For evaluation, we used the Dialog State Tracking Challenge (DSTC) 2 dataset - a popular belief tracking testbed with dialogs from restaurant information system. To our knowledge, our hybrid tracker sets a new state-of-the-art result in three out of four categories within the DSTC2. |
Year | Venue | DocType |
---|---|---|
2017 | EACL | Conference |
Volume | Citations | PageRank |
abs/1702.06336 | 1 | 0.36 |
References | Authors | |
12 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Miroslav Vodolán | 1 | 3 | 1.08 |
Rudolf Kadlec | 2 | 229 | 16.25 |
Jan Kleindienst | 3 | 220 | 23.74 |