'Long-Form Factually Correct Article Writer -- OpenAI GPT-3
I'm tasked with making a tool that writes articles based on either a question or an article title. For example: prompt: "top 10 reasons to move to Montana", response: [detailed article, 1500 words, factually accurate]
In theory, I could use the /answers endpoint and feed it a wikipedia article on a given topic (Montana) as the source of truth and GPT-3 would return a factually accurate, long-form article.
I haven't been able to get a response long enough, nor factually accurate enough. Any ideas? I suspect I'm not defining the "examples" and "examples_context" fields correctly.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|