[Idea] Categories of user queries

[Hansimov]

There are several typical categories of user queries. We could analyze them manually first, and then make this process automized and systemized.

Literal Facts

1. Keywords appeared in original texts (Easiest)

This kind of search task could be done by vector (or similarity) match via embedding.

Contexts example

Albert Einstein (14 March 1879 – 18 April 1955) was a German-born theoretical physicist who is widely held to be one of the greatest and most influential scientists of all time.
Best known for developing the theory of relativity, Einstein also made important contributions to quantum mechanics, and was thus a central figure in the revolutionary reshaping of the scientific understanding of nature that modern physics accomplished in the first decades of the twentieth century.

Queries example

1. When/Where was Einstein born?
2. Who is German theoretical physicist?
3. Tell me something about Einstein.
4. Who has contributed to quantum mechanics?

Structural Facts

1. Summary

Contexts example

(Same as above.)

Queries example

1. Summarize this paragraph.

See also

• LLM Benchmarks: MMLU, HellaSwag, BBH, and Beyond - Confident AI
  • https://www.confident-ai.com/blog/llm-benchmarks-mmlu-hellaswag-and-beyond