This AI can answer any questions you would have about me! 👋
App currently in development
Informations about me are divided into chunks
Each chunk of text is then embedded into a vector representation.
When you ask a question, this one is embedded in a similar way and the most similar chunk to the question in term of spatial representation is chosen.
If you want to know more the process, check these articles below:
Create your document chatbot with GPT-3 and LangChain
Semantic search using Sequence BERT
- Use an open-source LLM model
- Implement the chatbot into the website