O
Ondřej Dušek
Researcher at Charles University in Prague
Publications - 84
Citations - 2225
Ondřej Dušek is an academic researcher from Charles University in Prague. The author has contributed to research in topics: Natural language generation & Czech. The author has an hindex of 20, co-authored 84 publications receiving 1768 citations. Previous affiliations of Ondřej Dušek include Heriot-Watt University.
Papers
More filters
Proceedings ArticleDOI
Why We Need New Evaluation Metrics for NLG
TL;DR: A wide range of metrics are investigated, including state-of-the-art word-based and novel grammar-based ones, and it is demonstrated that they only weakly reflect human judgements of system outputs as generated by data-driven, end-to-end NLG.
Journal ArticleDOI
Evaluating the State-of-the-Art of End-to-End Natural Language Generation: The E2E NLG Challenge
TL;DR: The first shared task on end-to-end natural language generation (NLG) as discussed by the authors was the first attempt to compare machine learning and hand-engineered NLG models, and the results showed that seq2seq-based models generally score high in terms of word overlap metrics and human evaluations of naturalness.
Posted Content
The E2E Dataset: New Challenges For End-to-End Generation
TL;DR: The E2E dataset poses new challenges: (1) its human reference texts show more lexical richness and syntactic variation, including discourse phenomena; (2) generating from this set requires content selection, which promises more natural, varied and less template-like system utterances.
Proceedings ArticleDOI
Sequence-to-Sequence Generation for Spoken Dialogue via Deep Syntax Trees and Strings
Ondřej Dušek,Filip Jurčíček +1 more
TL;DR: This article presented a sequence-to-sequence approach that can be trained to produce natural language strings as well as deep syntax dependency trees from input dialogue acts, and they use it to directly compare two-step generation with separate sentence planning and surface realization stages to a joint, one-step approach.
Book ChapterDOI
CzEng 1.6: Enlarged Czech-English Parallel Corpus with Processing Tools Dockered
Ondřej Bojar,Ondřej Dušek,Tom Kocmi,Jindřich Libovický,Michal Novák,Martin Popel,Roman Sudarikov,Dusan Varis +7 more
TL;DR: The complete annotation pipeline as a virtual machine in the Docker virtualization toolkit is released, equipped with automatic annotation at a deep syntactic level of representation and alternatively in Universal Dependencies.