I’ve been looking around on hugging face for some pretrained models and I’ve found some models for text summarization and text classification. One of my plans was to have a text summarization model process articles and run a classification model to determine depth of an article to filter useless or garbage articles and save the useful ones to a feed.
I don’t know how successful this can be as especially foreign affairs news can be mostly hit pieces or propaganda, so if anyone has a better idea for this, or even knows how this fails, tips are appreciated!
I’m pretty new to LLMs, but have messed around with hugging face + transformers recently…