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

Attribute-based Controlled Text Generation (CTG) refers to generatingsentences that satisfy desirable attributes (e.g., emotions and topics).Existing works often utilize fine-tuning or resort to extra attributeclassifiers, yet suffer from storage and inference time increases. To addressthese concerns, we explore attribute-based CTG in a prompt-based manner. Inshort, the proposed Tailor represents each attribute as a pre-trainedcontinuous vector (i.e., single-attribute prompt) and guides the generation ofa fixed PLM switch to a pre-specified attribute. We experimentally find thatthese prompts can be simply concatenated as a whole to multi-attribute CTGwithout any re-training, yet raises problems of fluency decrease and positionsensitivity. To this end, Tailor provides a multi-attribute prompt mask and are-indexing position-ids sequence to bridge the gap between the training (oneprompt for each task) and testing stage (concatenating more than one prompt).To further enhance such single-attribute prompt combinations, Tailor alsointroduces a trainable prompt connector, which can be concatenated with any twosingle-attribute prompts to multi-attribute text generation. Experiments on 11attribute-specific generation tasks demonstrate strong performances of Tailoron both single-attribute and multi-attribute CTG, with 0.08\% trainingparameters of a GPT-2.

Tailor: A Prompt-Based Approach to Attribute-Based Controlled Text Generation