An alternative to using old school NLP, is to use GPT itself for the first pipeline as well, with a prompt like: I've the following resources with data. power_t

An alternative to using old school NLP, is to use GPT itself for the first pipeline as well, with a prompt like: I've the following resources with data. power_troubleshooting.txt contains information for customers that have issues powering on the device, (and so forth in the next lines... with other resources). This is the user question: ..., please reply with what is the resource I should access. Then you get the file and create a second prompt. Based on the following information: ..., answer this question: ... question ... A slower and more powerful way involves showing GPT different parts of potentially relevant text (for instance 3 each time) and ask it to score from 0 to 10 the level of usefulness of the resource in order to reply to the question. And make it select what resource to use. But this requires a lot of back and forth.