|
|
|
|
|
|
by karmakaze
669 days ago
|
|
|
The JSON representation is great for classical web scraping, but I wonder if LLMs are already good at 'seeing' HTML/CSS/js that it's more work to specialize. I could see Reddit as being a large source of content so worth the effort, at least until it's disabled--but that hasn't happened so far. |
|
|