K
Kellie Webster
Researcher at Google
Publications - 5
Citations - 225
Kellie Webster is an academic researcher from Google. The author has contributed to research in topics: Coreference & Sentiment analysis. The author has an hindex of 4, co-authored 5 publications receiving 107 citations. Previous affiliations of Kellie Webster include University of Sydney.
Papers
More filters
Proceedings ArticleDOI
Social Biases in NLP Models as Barriers for Persons with Disabilities
TL;DR: In this paper, the authors present evidence of such undesirable biases towards mentions of disability in two different English language models: toxicity prediction and sentiment analysis, and highlight topical biases in the discourse about disability which may contribute to the observed model biases.
Proceedings Article
Limited memory incremental coreference resolution
Kellie Webster,James Curran +1 more
TL;DR: An algorithm for coreference resolution based on analogy with shift-reduce parsing is proposed, which achieves CoNLL scores and is competitive with the best reported research systems despite having low memory requirements and a simpler model.
Posted Content
Social Biases in NLP Models as Barriers for Persons with Disabilities
TL;DR: Evidence of undesirable biases towards mentions of disability in two different English language models: toxicity prediction and sentiment analysis is presented and it is demonstrated that the neural embeddings that are the critical first step in most NLP pipelines similarly contain undesirable biases.
Proceedings ArticleDOI
Using Mention Accessibility to Improve Coreference Resolution
Kellie Webster,Joel Nothman +1 more
TL;DR: A novel fine-grained feature specialisation approach significantly improves the performance of a strong baseline, achieving state-of-the-art results of 65.29 and 61.13% on CoNLL-2012 using gold and automatic preprocessing.
Examining the Impact of Coreference Resolution on Quote Attribution
TL;DR: This work evaluates three quote attribution systems with automatically produced candidate speakers and coreference chains, and performs experiments over four separate corpora to determine how coreference resolution effects quote attribution, and to use the task as an extrinsic evaluation of three coreference systems.