Maximal Marginal Relevance - NISHIO Hirokazu's Scrapbox (Auto-translated from Japanese)

Maximal Marginal Relevance

Estimate a lower value for something similar to what you have already chosen.

https://gyazo.com/ed9cbd3c1e60fb8a9ed213ec862d4a8b

The use of MMR, diversity-based reranking for reordering documents and producing summaries

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.188.3982&rep=rep1&type=pdf

Maximal Marginal Relevance to Re-rank results in Unsupervised KeyPhrase Extraction | by Aditya Kumar | tech-that-works | Medium

[Linked from [Applying OpenAI's RAG Strategies https://blog.langchain.dev/applying-openai-rag/

Explanation of this Applying OpenAI's RAG strategy to LangChain｜npaka

Peripheral Relevance Maximization

An Attempt to Improve User Experience in Recommender Systems by Introducing Diversity (PDF) Yoshifumi Seki

I tried to apply MMR to the recommendation results to replace the rankings considering diversity │ Kiyoji's Proposition

---

This page is auto-translated from /nishio/Maximal Marginal Relevance using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I'm very happy to spread my thought to non-Japanese readers.