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

Lexically constrained sentence generation allows the incorporation of priorknowledge such as lexical constraints into the output. This technique has beenapplied to machine translation, and dialog response generation. Previous workusually used Markov Chain Monte Carlo (MCMC) sampling to generate lexicallyconstrained sentences, but they randomly determined the position to be editedand the action to be taken, resulting in many invalid refinements. To overcomethis challenge, we used a classifier to instruct the MCMC-based models whereand how to refine the candidate sentences. First, we developed two methods tocreate synthetic data on which the pre-trained model is fine-tuned to obtain areliable classifier. Next, we proposed a two-step approach, "Predict andRevise", for constrained sentence generation. During the predict step, weleveraged the classifier to compute the learned prior for the candidatesentence. During the revise step, we resorted to MCMC sampling to revise thecandidate sentence by conducting a sampled action at a sampled position drawnfrom the learned prior. We compared our proposed models with many strongbaselines on two tasks, generating sentences with lexical constraints and textinfilling. Experimental results have demonstrated that our proposed modelperforms much better than the previous work in terms of sentence fluency anddiversity. Our code and pre-trained models are available athttps://github.com/NLPCode/MCMCXLNet.

Show Me How To Revise: Improving Lexically Constrained Sentence Generation with XLNet