English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

While both extractive and generative readers have been successfully appliedto the Question Answering (QA) task, little attention has been paid toward thesystematic comparison of them. Characterizing the strengths and weaknesses ofthe two readers is crucial not only for making a more informed reader selectionin practice but also for developing a deeper understanding to foster furtherresearch on improving readers in a principled manner. Motivated by this goal,we make the first attempt to systematically study the comparison of extractiveand generative readers for question answering. To be aligned with thestate-of-the-art, we explore nine transformer-based large pre-trained languagemodels (PrLMs) as backbone architectures. Furthermore, we organize our findingsunder two main categories: (1) keeping the architecture invariant, and (2)varying the underlying PrLMs. Among several interesting findings, it isimportant to highlight that (1) the generative readers perform better in longcontext QA, (2) the extractive readers perform better in short context whilealso showing better out-of-domain generalization, and (3) the encoder ofencoder-decoder PrLMs (e.g., T5) turns out to be a strong extractive reader andoutperforms the standard choice of encoder-only PrLMs (e.g., RoBERTa). We alsostudy the effect of multi-task learning on the two types of readers varying theunderlying PrLMs and perform qualitative and quantitative diagnosis to providefurther insights into future directions in modeling better readers.

Choose Your QA Model Wisely: A Systematic Study of Generative and Extractive Readers for Question Answering