search

madaan / self-refine

LLMs can generate feedback on their work, use it to improve the output, and repeat this process iteratively.
https://selfrefine.info
Apache License 2.0
596 stars 48 forks source link

Demo doesn't seem to work #14

Open ogencoglu opened 1 year ago

ogencoglu commented 1 year ago

There is never a feedback

madaan commented 1 year ago

Unfortunately, I ran out of OpenAI credits to support the demo. I'll add an option of plugging in a key soon-ish.