If you have been following me for a while, you might have heard of the "AI search loop." Given a question, this method is a thorough yet rapid way to find relevant scientific literature. Today's newsletter covers recent updates to all involved tools. Most interestingly, Google Scholar is now becoming an alternative to pricy AI tools. (See below what it means for the big picture.)
Here's how the search loop works: Based on a plain-text question, you use Consensus or the new Google Scholar Labs AI feature to find a few seed papers. These results are highly targeted, but typically yield only 2-3 relevant papers. This is why in the next step, you use Litmaps or ResearchRabbit to dig into the references of these first few papers to find additional literature based on the reference network (i.e. who cites whom). The combination of AI and reference network is what makes it so powerful. Here is a schematic:
No comments:
Post a Comment