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 sequence-to-sequence models have shown remarkable generalization poweracross several natural language tasks, their construct of solutions are arguedto be less compositional than human-like generalization. In this paper, wepresent seq2attn, a new architecture that is specifically designed to exploitattention to find compositional patterns in the input. In seq2attn, the twostandard components of an encoder-decoder model are connected via a transcoder,that modulates the information flow between them. We show that seq2attn cansuccessfully generalize, without requiring any additional supervision, on twotasks which are specifically constructed to challenge the compositional skillsof neural networks. The solutions found by the model are highly interpretable,allowing easy analysis of both the types of solutions that are found andpotential causes for mistakes. We exploit this opportunity to introduce a newparadigm to test compositionality that studies the extent to which a modelovergeneralizes when confronted with exceptions. We show that seq2attn exhibitssuch overgeneralization to a larger degree than a standard sequence-to-sequencemodel.

Transcoding compositionally: using attention to find more generalizable solutions