News for hackers. Stuff that matters.
Re:RAGFlow is an open-source RAG engine based on OCR and doc... by constantinum on Monday April 01, 2024 @10:43PM (#39901877)
Document processing is getting better and better with new tools leveraging LLMs. If anyone is interested in exploring this space, try another similar tool LLMWhisperer (https://llmwhisperer.unstract.com/). It is a part of Unstract, an open-source document processing tool (https://github.com/Zipstack/unstract)
Re:RAGFlow is an open-source RAG engine based on OCR and doc... by TuringNYC on Tuesday April 02, 2024 @12:29AM (#39902392)
Do you have any recommendations for extraction from Powerpoint documents? Those seem like the worst since each of the layouts tend to be unique (unlike the AMEX Statements in the example)
Re:RAGFlow is an open-source RAG engine based on OCR and doc... by demilich on Tuesday April 02, 2024 @05:30AM (#39903817)
Cool! This is really helpful.
Re:RAGFlow is an open-source RAG engine based on OCR and doc... by yingfeng on Tuesday April 02, 2024 @05:48AM (#39903903)
Actually we've tried almost all lof existing open source models for document processing, and none of them performs well for complex documents, especially those having complicated tables, such as tables cells without borders, cells need to be combined,...,etc. Although adopting LLMs to perform such document understanding tasks is more scalable, it requires much more data and computation power to achieve similar results. That's why we design such models start from scratch.