Hacker News Reader

The idea is that Unsiloed would check for values that are hard to read and Jev checks for semantic issues. No one is prescribing 10 tablets, 3 times a day. In this example it's Jev's job to check for obvious errors in units and so it flags this value. Unsiloed cannot flag it because, as you said, the text says exactly that. That's why the models work well together as they catch different issues.

View on HN

But you need to explicitly specify the prompt to check dosage. If you passed it as part of the original extraction model call, it would return a lower confidence to begin with. Or provide a different 'is_dosage_reasonable' value, which would eliminate the need for Jev.

And this is before one argues that dose <= 3 tablets is a better validation than the Jev prompt and costs nothing.

Well yes but then you'd have to manually code every set of expectations, the point of using a model like Jev is it can flag a "something is wrong here" entry without needing someone to define every thing that could go wrong

Reply on news.ycombinator.com