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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Abstractive summarization, the task of generating a concise summary of inputdocuments, requires: (1) reasoning over the source document to determine thesalient pieces of information scattered across the long document, and (2)composing a cohesive text by reconstructing these salient facts into a shortersummary that faithfully reflects the complex relations connecting these facts.In this paper, we adapt TP-TRANSFORMER (Schlag et al., 2019), an architecturethat enriches the original Transformer (Vaswani et al., 2017) with theexplicitly compositional Tensor Product Representation (TPR), for the task ofabstractive summarization. The key feature of our model is a structural biasthat we introduce by encoding two separate representations for each token torepresent the syntactic structure (with role vectors) and semantic content(with filler vectors) separately. The model then binds the role and fillervectors into the TPR as the layer output. We argue that the structuredintermediate representations enable the model to take better control of thecontents (salient facts) and structures (the syntax that connects the facts)when generating the summary. Empirically, we show that our TP-TRANSFORMERoutperforms the Transformer and the original TP-TRANSFORMER significantly onseveral abstractive summarization datasets based on both automatic and humanevaluations. On several syntactic and semantic probing tasks, we demonstratethe emergent structural information in the role vectors and improved syntacticinterpretability in the TPR layer outputs. Code and models are available athttps://github.com/jiangycTarheel/TPT-Summ.

Enriching Transformers with Structured Tensor-Product Representations for Abstractive Summarization