What are the most effective recent approaches for predicting social media post virality?

I’m currently working on a project related to virality prediction . I came across this 2024 paper that combines BERT and CNN for Twitter virality classification:
:link: Virality Prediction on Twitter Using Combined CNN and BERT Models | IEEE Xplore

Do you think this BERT+CNN hybrid is a good choice in 2024/2025?
Are there more advanced or better-performing models (e.g. graph-based, transformer-only, multimodal) that you’d recommend for this task?

Any suggestions or insights from your experience would be greatly appreciated!

1 Like

I can’t find any methods other than BERT-based models…

https://www.researchgate.net/publication/355473219_Virality_Prediction_for_News_Tweets_Using_RoBERTa