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

Hallucination of text ungrounded in the input is a well-known problem inneural data-to-text generation. Many methods have been proposed to mitigate it,but they typically require altering model architecture or collecting additionaldata, and thus cannot be easily applied to an existing model. In this paper, weexplore a new way to mitigate hallucinations by combining the probabilisticoutput of a generator language model (LM) with the output of a special "textcritic" classifier, which guides the generation by assessing the match betweenthe input data and the text generated so far. Our method does not need anychanges to the underlying LM's architecture or training procedure and can thusbe combined with any model and decoding operating on word probabilities. Thecritic does not need any additional training data, using the base LM's trainingdata and synthetic negative examples. Our experimental results show that ourmethod improves over the baseline on the WebNLG and OpenDialKG benchmarks.

Critic-Driven Decoding for Mitigating Hallucinations in Data-to-text Generation