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

Controllable text generation is a challenging and meaningful field in naturallanguage generation (NLG). Especially, poetry generation is a typical one withwell-defined and strict conditions for text generation which is an idealplayground for the assessment of current methodologies. While prior workssucceeded in controlling either semantic or metrical aspects of poetrygeneration, simultaneously addressing both remains a challenge. In this paper,we pioneer the use of the Diffusion model for generating sonnets and ChineseSongCi poetry to tackle such challenges. In terms of semantics, ourPoetryDiffusion model, built upon the Diffusion model, generates entiresentences or poetry by comprehensively considering the entirety of sentenceinformation. This approach enhances semantic expression, distinguishing it fromautoregressive and large language models (LLMs). For metrical control, theseparation feature of diffusion generation and its constraint control moduleenable us to flexibly incorporate a novel metrical controller to manipulate andevaluate metrics (format and rhythm). The denoising process in PoetryDiffusionallows for gradual enhancement of semantics and flexible integration of themetrical controller which can calculate and impose penalties on states thatstray significantly from the target control distribution. Experimental resultson two datasets demonstrate that our model outperforms existing models inautomatic evaluation of semantic, metrical, and overall performance as well ashuman evaluation.

PoetryDiffusion: Towards Joint Semantic and Metrical Manipulation in Poetry Generation