We wonder that can we generate sentence to prompt using different LLM like GPT4. Then change couple words in that sentence using MASK model so as a last result sentence is going to be converted to persona's style.
Steps
[ ] Fine tune a text model
[ ] Test that model using couple sentences and observe how does it works
[ ] Report the results. If this makes more sentence than QA model logic we can use this
Goal
We wonder that can we generate sentence to prompt using different LLM like GPT4. Then change couple words in that sentence using MASK model so as a last result sentence is going to be converted to persona's style.
Steps