Get the latest tech news
Coping with dumb LLMs using classic ML
Taking the output of many dumb LLM search relevance judges and feeding the output to a decision tree to improve precision
My goal, not so much to replace other labels but to at least be a reliable to flag what looks amiss / promising much faster without needing to always recruit humans. But the upshot of all this is we can take lots of dumb agents, making single, basic decisions, and combine their outputs into something smarter using traditional ML. We keep their decisions dumb, simple, and interpretable combining them at the end with fast boring, old ML that can finish the job.
Or read this on Hacker News