PixelRAG beats text parsers on accuracy and cuts AI agent token costs 10x

UC Berkeley's PixelRAG renders pages as screenshots instead of parsing text, boosting RAG accuracy by up to 18.1% and cutting AI agent token costs 10x.

Mistral OCR 4 with structured document extraction, 170 languages and self-hosting launched

Mistral AI has introduced Mistral OCR 4, a new optical character recognition (OCR) model designed for enterprise document ...
Tech Times

Mistral OCR 4 Ships: Structure-Aware Document AI Runs On Your Own Infrastructure

Mistral OCR 4 brings bounding boxes, typed-block classification, and 170-language document extraction to enterprises that ...
VentureBeat

AWS debuts advanced RAG features for structured, unstructured data

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Getting enterprise data into large language models (LLMs) is a critical ...
Forbes

Retrieval-Augmented Generation Is An Engineering Problem, Not A Model Problem

Every few months, the enterprise AI conversation resets around the same flawed premise that better models solve the problem. When large language models hallucinate, the instinct is to reach for a ...
Law

Questel Unveils AI Retrieval Model for Patent Tools

QaECTER is an embedding model that performs patent retrieval and is designed to improve the output of Questel’s gen AI products. QaECTER powers an upgrade to semantic search across Questel's product ...