Skip to content

Prompt Engineered GPT-4 Beats Gemini on all of Google's text benchmarks

microsoft.com
21 pointsFlyingLawnmower3 comments
On HN

Comments

"We note that Medprompt+ relies on accessing confidence scores (logprobs) from GPT-4. These are not publicly available via the current API but will be enabled for all in the near future."

Had that been announced yet?

It would be interesting to see how this can be used for open models.

engineered? Over overfit?

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.