Commonly used algorithms to compare texts
Hi! I'm new to computational linguistics and recently I need to estimate how much of a text our participants can remember for a project. So far we had a list of "information units" that are in the text, and we manually checked if the participants mentioned them in what they wrote. Now we want to automate this process. I tried to look for machine learning approaches, but I found mostly sentiment analysis papers or word counts, plus a lot with LLMs (however the latter didn't look very standard in the field to me, more like a new approach). Also, algorithms you have to train, but we don't have enough data to do so. In general there was a lot, so I had trouble knowing what to choose or where to even start.
Is there any algorithm or tool already trained that is commonly used for this? Any insights or guidance is appreciated.