Log in to save to my catalogue

Robust Dialogue State Tracking with Weak Supervision and Sparse Data

Robust Dialogue State Tracking with Weak Supervision and Sparse Data

https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_proquest_journals_2893948503

Robust Dialogue State Tracking with Weak Supervision and Sparse Data

About this item

Full title

Robust Dialogue State Tracking with Weak Supervision and Sparse Data

Publisher

One Broadway, 12th Floor, Cambridge, Massachusetts 02142, USA: MIT Press

Journal title

Transactions of the Association for Computational Linguistics, 2022-11, Vol.10, p.1175-1192

Language

English

Formats

Publication information

Publisher

One Broadway, 12th Floor, Cambridge, Massachusetts 02142, USA: MIT Press

More information

Scope and Contents

Contents

Generalizing dialogue state tracking (DST) to new data is especially challenging due to the strong reliance on abundant and fine-grained supervision during training. Sample sparsity, distributional shift, and the occurrence of new concepts and topics frequently lead to severe performance degradation during inference. In this paper we propose a trai...

Alternative Titles

Full title

Robust Dialogue State Tracking with Weak Supervision and Sparse Data

Identifiers

Primary Identifiers

Record Identifier

TN_cdi_proquest_journals_2893948503

Permalink

https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_proquest_journals_2893948503

Other Identifiers

ISSN

2307-387X

E-ISSN

2307-387X

DOI

10.1162/tacl_a_00513

How to access this item