Jul 10, 2021 · We introduce a four-stage protocol which gives an approximate measure of how much in-domain, "target" data a parser requires to achieve a certain quality bar.

Data efficiency, despite being an attractive characteristic, is often challenging to measure and optimize for in task-oriented semantic parsing; unlike exact ...

Data efficiency, despite being an attractive characteristic, is often challenging to measure and optimize for in task-oriented semantic parsing; ...

As part of this work we explore leveraging external knowledge to improve model accuracy in low-resource and low-compute settings.

RETRONLU: Retrieval Augmented Task-Oriented Semantic Parsing · Assessing Data Efficiency in Task-Oriented Semantic Parsing · Latency-Aware Neural Architecture ...

Data efficiency, despite being an attractive characteristic, is often challenging to measure and optimize for in task-oriented semantic parsing; unlike exact ...

Feb 15, 2023 · We conclude that for decoder initialization, having some intelligent initializa- tion obtained from ASR pre-training is adequate and does not.

Data efficiency, despite being an attractive characteristic, is often challenging to measure and optimize for in task-oriented semantic parsing; unlike exact ...

We perform an extensive evaluation of deep-learning techniques for task-oriented parsing on this dataset, including different flavors of seq2seq systems and ...

Abstract. We introduce Recursive INsertion-based Encoder (RINE), a novel approach for semantic parsing in task-oriented dialog.