A toolkit for training language models to work with PDF documents in the wild, including prompting strategies, evaluation tools, filtering, finetuning code, and processing PDFs through finetuned models.
The article discusses the process of preparing PDFs for use in Retrieval-Augmented Generation (RAG) systems, with a focus on creating graph-based RAGs from annual reports containing tables. It highlights the benefits of Graph RAGs over vector store-backed RAGs, particularly in terms of reasoning capabilities, and explores the construction of knowledge graphs for better information retrieval. The author shares insights into the challenges and solutions involved in building an enterprise-ready graph data store for RAG applications.
MarkItDown is a utility for converting various files to Markdown, including PDF, PowerPoint, Word, Excel, Images, Audio, HTML, text-based formats, and ZIP files.
How to read and convert PDFs to Markdown for better RAG results with LLMs.
Docling is a tool that parses documents and exports them to desired formats like Markdown and JSON. It supports various document formats including PDF, DOCX, PPTX, Images, HTML, AsciiDoc, and Markdown.
- Approximate Tokens, Words and Characters Calculator for LLM's and Text Trimmer — Simple calculator to estimate tokens for Large Language Models and text editor to trim text
- Text File Merger for LLM — This tool combines multiple text files into a single document, with clear separation between files
- PDF to TXT Converter — Convert PDF documents to plain text format for use with LLMs and text analysis
- HTML to TXT Converter — Remove HTML tags and extract clean text content for LLM processing
- LLM System Prompt Generator — Generate optimized system prompts for different LLM model sizes (3B, 33B, 70B, etc.)
- Creative Idea Generator — AI-powered brainstorming tool for generating creative solutions and ideas
This blog post explores scaling ColPali for efficient document retrieval across large collections of PDFs using Vespa's phased retrieval and ranking pipeline, including the use of a hamming-based MaxSim similarity function.
IncarnaMind enables chatting with personal documents (PDF, TXT) using Large Language Models (LLMs) like GPT. It uses a Sliding Window Chunking mechanism and Ensemble Retriever for efficient querying.
A post discussing new techniques developed for parsing and searching PDFs, focusing on turning them into a hierarchical structure for RAG search. The approach involves dynamically generating chunks for searches, sending headers and sub-headers to the Language Model along with relevant chunks.
The llmsherpa project provides APIs to accelerate Large Language Model (LLM) projects. It includes features like LayoutPDFReader for PDF text parsing, smart chunking for vector search and Retrieval Augmented Generation, and table analysis. It is open-sourced under Apache 2.0 license.
