Post

RAG papers list newest additions

๐ŸŽŠ My RAG papers list has now been updated with the newest additions, including those from April. It contains over 60 papers, with quick summaries and topic tags.

๐Ÿ”‰ Over the past year, the release of new LLMs and their increasing application across various fields has also spurred a surge in research on RAG approaches. A ton of advanced methods have been proposed to boost RAGโ€™s efficiency in step with the wider adoption of LLMs.

๐Ÿ’ก I have compiled a selection of the most popular papers on RAG starting from April 2023 to the present, categorized as follows:

โ›ณ RAG Survey ๐Ÿ“– Comprehensive overview of existing methods in RAG.

โ›ณ RAG Enhancement (Advanced Techniques) ๐Ÿ“– Proposals for improving the efficiency and effectiveness of the RAG pipeline.

โ›ณ Retrieval Improvement ๐Ÿ“– Techniques focused on enhancing the retrieval component of RAG.

โ›ณ Comparison Papers ๐Ÿ“– Papers comparing RAG with other methods or approaches.

โ›ณ Domain-Specific RAG ๐Ÿ“–Adaptation of RAG techniques for specific domains or applications.

โ›ณRAG Evaluation: ๐Ÿ“–Assessment of the performance and effectiveness of RAG models.

โ›ณRAG Embeddings: ๐Ÿ“–Methods for developing better embedding techniques optimized for RAG or retrieval in RAG.

โ›ณInput Processing for RAG: ๐Ÿ“–Techniques for preprocessing input data to optimize the performance and effectiveness of RAG models.

Depending on the use-case, you can explore relevant papers to address various challenges and improve RAG. Happy Learning!

Link to the list: https://github.com/aishwaryanr/awesome-generative-ai-guide/blob/main/research_updates/rag_research_table.md

๐Ÿšจ I post #genai content daily, follow along for the latest updates #rag #llms

 RAG Papers newest additions

Translate to Korean

๐ŸŽŠ ๋‚ด RAG ๋…ผ๋ฌธ ๋ชฉ๋ก์€ ์ด์ œ 4์›”์˜ ๋…ผ๋ฌธ์„ ํฌํ•จํ•˜์—ฌ ์ตœ์‹  ์ถ”๊ฐ€ ์‚ฌํ•ญ์œผ๋กœ ์—…๋ฐ์ดํŠธ๋˜์—ˆ์Šต๋‹ˆ๋‹ค. ์—ฌ๊ธฐ์—๋Š” ๋น ๋ฅธ ์š”์•ฝ๊ณผ ์ฃผ์ œ ํƒœ๊ทธ์™€ ํ•จ๊ป˜ 60๊ฐœ ์ด์ƒ์˜ ๋…ผ๋ฌธ์ด ํฌํ•จ๋˜์–ด ์žˆ์Šต๋‹ˆ๋‹ค.

๐Ÿ”‰ ์ง€๋‚œ ํ•œ ํ•ด ๋™์•ˆ ์ƒˆ๋กœ์šด LLM์ด ์ถœ์‹œ๋˜๊ณ  ๋‹ค์–‘ํ•œ ๋ถ„์•ผ์—์„œ LLM์˜ ์ ์šฉ์ด ์ฆ๊ฐ€ํ•จ์— ๋”ฐ๋ผ RAG ์ ‘๊ทผ ๋ฐฉ์‹์— ๋Œ€ํ•œ ์—ฐ๊ตฌ๊ฐ€ ๊ธ‰์ฆํ–ˆ์Šต๋‹ˆ๋‹ค. LLM์˜ ๊ด‘๋ฒ”์œ„ํ•œ ์ฑ„ํƒ์— ๋ฐœ๋งž์ถฐ RAG์˜ ํšจ์œจ์„ฑ์„ ๋†’์ด๊ธฐ ์œ„ํ•ด ์ˆ˜๋งŽ์€ ๊ณ ๊ธ‰ ๋ฐฉ๋ฒ•์ด ์ œ์•ˆ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.

๐Ÿ’ก 2023๋…„ 4์›”๋ถ€ํ„ฐ ํ˜„์žฌ๊นŒ์ง€ RAG์— ๋Œ€ํ•œ ๊ฐ€์žฅ ์ธ๊ธฐ ์žˆ๋Š” ๋…ผ๋ฌธ์„ ๋‹ค์Œ๊ณผ ๊ฐ™์ด ๋ถ„๋ฅ˜ํ•˜์—ฌ ํŽธ์ง‘ํ–ˆ์Šต๋‹ˆ๋‹ค.

โ›ณ RAG ์„ค๋ฌธ์กฐ์‚ฌ ๐Ÿ“– RAG์˜ ๊ธฐ์กด ๋ฐฉ๋ฒ•์— ๋Œ€ํ•œ ํฌ๊ด„์ ์ธ ๊ฐœ์š”.

โ›ณ RAG ํ–ฅ์ƒ(๊ณ ๊ธ‰ ๊ธฐ์ˆ ) ๐Ÿ“– RAG ํŒŒ์ดํ”„๋ผ์ธ์˜ ํšจ์œจ์„ฑ๊ณผ ํšจ๊ณผ๋ฅผ ๊ฐœ์„ ํ•˜๊ธฐ ์œ„ํ•œ ์ œ์•ˆ.

โ›ณ ๊ฒ€์ƒ‰ ๊ฐœ์„  ๐Ÿ“– RAG์˜ ๊ฒ€์ƒ‰ ๊ตฌ์„ฑ ์š”์†Œ๋ฅผ ํ–ฅ์ƒ์‹œํ‚ค๋Š” ๋ฐ ์ค‘์ ์„ ๋‘” ๊ธฐ์ˆ .

โ›ณ ๋น„๊ต ๋…ผ๋ฌธ ๐Ÿ“– RAG๋ฅผ ๋‹ค๋ฅธ ๋ฐฉ๋ฒ• ๋˜๋Š” ์ ‘๊ทผ ๋ฐฉ์‹๊ณผ ๋น„๊ตํ•˜๋Š” ๋…ผ๋ฌธ.

โ›ณ ๋„๋ฉ”์ธ๋ณ„ RAG ๐Ÿ“–ํŠน์ • ๋„๋ฉ”์ธ ๋˜๋Š” ์‘์šฉ ํ”„๋กœ๊ทธ๋žจ์— ๋Œ€ํ•œ RAG ๊ธฐ์ˆ  ์ ์šฉ.

โ›ณRAG ํ‰๊ฐ€: ๐Ÿ“–RAG ๋ชจ๋ธ์˜ ์„ฑ๋Šฅ ๋ฐ ํšจ๊ณผ ํ‰๊ฐ€.

โ›ณRAG ์ž„๋ฒ ๋”ฉ: ๐Ÿ“–RAG์—์„œ RAG ๋˜๋Š” ๊ฒ€์ƒ‰์— ์ตœ์ ํ™”๋œ ๋” ๋‚˜์€ ์ž„๋ฒ ๋”ฉ ๊ธฐ์ˆ ์„ ๊ฐœ๋ฐœํ•˜๋Š” ๋ฐฉ๋ฒ•.

โ›ณRAG์— ๋Œ€ํ•œ ์ž…๋ ฅ ์ฒ˜๋ฆฌ: ๐Ÿ“–RAG ๋ชจ๋ธ์˜ ์„ฑ๋Šฅ๊ณผ ํšจ์œจ์„ฑ์„ ์ตœ์ ํ™”ํ•˜๊ธฐ ์œ„ํ•ด ์ž…๋ ฅ ๋ฐ์ดํ„ฐ๋ฅผ ์ „์ฒ˜๋ฆฌํ•˜๋Š” ๊ธฐ์ˆ ์ž…๋‹ˆ๋‹ค.

์‚ฌ์šฉ ์‚ฌ๋ก€์— ๋”ฐ๋ผ ๊ด€๋ จ ๋ฌธ์„œ๋ฅผ ํƒ์ƒ‰ํ•˜์—ฌ ๋‹ค์–‘ํ•œ ๋ฌธ์ œ๋ฅผ ํ•ด๊ฒฐํ•˜๊ณ  RAG๋ฅผ ๊ฐœ์„ ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ํ–‰๋ณตํ•œ ํ•™์Šต!

๋ชฉ๋ก์— ๋Œ€ํ•œ ๋งํฌ: https://github.com/aishwaryanr/awesome-generative-ai-guide/blob/main/research_updates/rag_research_table.md

This post is licensed under CC BY 4.0 by the author.