Findings of the E2E NLG Challenge
This work addresses the problem of limited data annotation needs for researchers and developers in natural language processing, but it is incremental as it builds on existing shared task frameworks.
The paper tackled the challenge of improving end-to-end natural language generation in spoken dialogue systems by evaluating novel approaches on a dataset with higher lexical richness and syntactic complexity, finding that 62 systems from 17 institutions were compared, with most using sequence-to-sequence models.
This paper summarises the experimental setup and results of the first shared task on end-to-end (E2E) natural language generation (NLG) in spoken dialogue systems. Recent end-to-end generation systems are promising since they reduce the need for data annotation. However, they are currently limited to small, delexicalised datasets. The E2E NLG shared task aims to assess whether these novel approaches can generate better-quality output by learning from a dataset containing higher lexical richness, syntactic complexity and diverse discourse phenomena. We compare 62 systems submitted by 17 institutions, covering a wide range of approaches, including machine learning architectures -- with the majority implementing sequence-to-sequence models (seq2seq) -- as well as systems based on grammatical rules and templates.