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

The principle of the Information Bottleneck (Tishby et al. 1999) is toproduce a summary of information X optimized to predict some other relevantinformation Y. In this paper, we propose a novel approach to unsupervisedsentence summarization by mapping the Information Bottleneck principle to aconditional language modelling objective: given a sentence, our approach seeksa compressed sentence that can best predict the next sentence. Our iterativealgorithm under the Information Bottleneck objective searches gradually shortersubsequences of the given sentence while maximizing the probability of the nextsentence conditioned on the summary. Using only pretrained language models withno direct supervision, our approach can efficiently perform extractive sentencesummarization over a large corpus.  Building on our unsupervised extractive summarization (BottleSumEx), we thenpresent a new approach to self-supervised abstractive summarization(BottleSumSelf), where a transformer-based language model is trained on theoutput summaries of our unsupervised method. Empirical results demonstrate thatour extractive method outperforms other unsupervised models on multipleautomatic metrics. In addition, we find that our self-supervised abstractivemodel outperforms unsupervised baselines (including our own) by humanevaluation along multiple attributes.

BottleSum: Unsupervised and Self-supervised Sentence Summarization using the Information Bottleneck Principle