zlacker

[return to "Pushing ChatGPT's Structured Data Support to Its Limits"]
1. JoshMa+Dg[view] [source] 2023-12-27 16:42:33
>>goranm+(OP)
FWIW, I've seen stronger performance from gpt-4-1106-preview when I use `response_format: { type: "json_object" },` (providing a target typescript interface in context), vs the "tools" API.

More flexible, and (evaluating non-scientifically!) qualitatively better answers & instruction following -- particularly for deeply nested or complex schemas, which typescript expresses very clearly and succinctly.

Example from a hack week project earlier this month (using a TS-ish schema description that's copy/pasted from healthcare's FHIR standard): https://github.com/microsoft-healthcare-madison/hackweek-202...

Or a more complex example with one model call to invent a TS schema on-the-fly and another call to abstract clinical data into it: https://github.com/microsoft-healthcare-madison/hackweek-202...

◧◩
2. joshua+eq[view] [source] 2023-12-27 17:35:40
>>JoshMa+Dg
I still have a lot of trouble using OpenAI function-calling API (and json mode) for returning source code. I have trouble when the source code has quotes, which is pretty common. The result may be missing quotes, have incorrectly escaped quotes, or use of the wrong type of quotes at the JSON object level.
◧◩◪
3. visarg+Iy[view] [source] 2023-12-27 18:21:30
>>joshua+eq
Same for YAML, you got to be careful with strings
[go to top]