10 ms·
> I’ve seen LLMs find matches for queries like this, even if “Dick” is a nickname for “Richard”. An awful lot of business questions are in this category and nei
by drpixie 3y ago
> I’ve seen LLMs find matches for queries like this, even if “Dick” is a nickname for “Richard”. An awful lot of business questions are in this category and neither relational theory nor full text search is useful to solve such problems.
Question - have you seen an LLM that can answer queries like this correctly and consistently, for a range of natural inputs?
The ones I've seen often produce correct answers, but likely often produce apparently correct but wrong answers, and often produce garbage. For those systems, we've just changed the human's work from "write a query" to "filter the correct answers from the garbage" :(
- idopmstuff 3y agoYes, provided the answer for the question is in context, not training. If you feed GPT4 a document and ask questions, it will virtually always answer those questions right. If you ask it for info in its training set, it will hallucinate sometimes. With Anthropic's 100k token context window, you can do a whole lot with this.