Emojich -- zero-shot emoji generation using Russian language: a technical report
This work addresses the need for emoji generation in Russian language, but it is incremental as it builds on an existing pretrained model with engineering tweaks.
The authors tackled the problem of generating emojis from Russian text captions by fine-tuning the ruDALL-E Malevich model, resulting in a system that produces customized emoji sets with preserved generalization and a specific style.
This technical report presents a text-to-image neural network "Emojich" that generates emojis using captions in Russian language as a condition. We aim to keep the generalization ability of a pretrained big model ruDALL-E Malevich (XL) 1.3B parameters at the fine-tuning stage, while giving special style to the images generated. Here are presented some engineering methods, code realization, all hyper-parameters for reproducing results and a Telegram bot where everyone can create their own customized sets of stickers. Also, some newly generated emojis obtained by "Emojich" model are demonstrated.