LongTracer: Open-Source RAG Hallucination Detection Without LLM-as-a-Judge
Stop paying to evaluate your LLM outputs. Stop tolerating non-deterministic quality gates. LongTracer...
#RAG
1132 articles tagged with RAG
LLM Fine-Tuning vs RAG: A Production Decision Framework for Engineering Teams
Key Takeaways Use RAG for knowledge retrieval, changing data, and rapid iteration. Use fine-tuning...
Как и зачем мы сделали собственный OCR-бенчмаркОднажды нам понадобилось выбрать OCR-модель для RAG-пайплайна. Казалось б...
Как и зачем мы сделали собственный OCR-бенчмаркОднажды нам понадобилось выбрать OCR-модель для RAG-пайплайна. Казалось бы, задача простая: смотришь на лидерборды, берешь лучшую, PR...
Building a production RAG across a Book series: Retrieval, Reranking, and Hard Lessons
I built a search and Q&A system over the entire A Song of Ice and Fire series, all 10 books,...
RAG Evaluation Checklist for AI SaaS: Catch Bad Answers Before Users Do
A practical RAG evaluation checklist for AI SaaS builders to test retrieval quality, grounded answers, citations, regressions, and production failure replay.
Answer Presence Drives RAG Rewriting Gains
Retrieval-augmented QA pipelines often route retrieved passages through an LLM rewriter before a smaller reader, lifting F1 by tens of points on multi-hop benchmarks; this gain is ...
Artain-AI/ignite-ms: Fast self-hosted embedding engine for search, RAG, and reindexing workloads on NVIDIA GPUs. Built in Rust + TensorRT for teams that care about scale, cost, and control.
Fast self-hosted embedding engine for search, RAG, and reindexing workloads on NVIDIA GPUs. Built in Rust + TensorRT for teams that care about scale, cost, and control.
My custom rag build on #axelera's preorder isn't going great.... 4gb aipu is too little..That, and also I cannot use #ll...
My custom rag build on #axelera's preorder isn't going great.... 4gb aipu is too little..That, and also I cannot use #llm from huggingface because they won't axcompile.I feel bad r...
# Enterprise RAG’s Biggest Risk: Answers That Look Correct but Aren’t
Most RAG demos feel impressive at first. You upload documents, ask a question, and the system...
How we index images for RAGhttps://www.kapa.ai/blog/how-we-index-images-for-rag#ai
How we index images for RAGhttps://www.kapa.ai/blog/how-we-index-images-for-rag#ai
RAG reranking for production agents: four approaches, four failure modes
Most agents that "hallucinate" in production aren't actually hallucinating. The right context existed...
I Work in Healthcare Tech. Here's Why I Built a RAG Tool for Clinical Documents.
I didn't set out to build a RAG application. I set out to solve an annoying problem I kept watching...
💡 What makes a RAG system actually production-ready?Not the LLM. Not even the vector DB.It's the CHUNKING strategy:📄 Fix...
💡 What makes a RAG system actually production-ready?Not the LLM. Not even the vector DB.It's the CHUNKING strategy:📄 Fixed-size: Fast, dumb📄 Semantic: Slow, smart📄 Hierarchical: Be...
Гибридный поиск в RAG: как мы подняли Top-1 с 62% до 88% на базе из 50 000 документовЕсли вы строили RAG, вы знаете эту ...
Гибридный поиск в RAG: как мы подняли Top-1 с 62% до 88% на базе из 50 000 документовЕсли вы строили RAG, вы знаете эту боль: вектор-поиск красиво работает на демо-вопросах, но в п...
How we index images for RAGhttps://www.kapa.ai/blog/how-we-index-images-for-rag#HackerNews #imageindexing #RAG #machinel...
How we index images for RAGhttps://www.kapa.ai/blog/how-we-index-images-for-rag#HackerNews #imageindexing #RAG #machinelearning #AI #blogpost #technology
로컬 AI와 RAG의 한계: 모델의 본성과 '원죄'에 대하여 로컬 AI 모델은 학습 데이터에 의해 결정되는 '원죄'를 가지며, 학습되지 않은 특정 도메인(Ruby-LibGD 등)에서는 근본적인 지능의 한계가 존재합니...
로컬 AI와 RAG의 한계: 모델의 본성과 '원죄'에 대하여 로컬 AI 모델은 학습 데이터에 의해 결정되는 '원죄'를 가지며, 학습되지 않은 특정 도메인(Ruby-LibGD 등)에서는 근본적인 지능의 한계가 존재합니다. 🔗 원문 보기
How we index images for RAGhttps://www.kapa.ai/blog/how-we-index-images-for-rag#HackerNews #Tech #AI
How we index images for RAGhttps://www.kapa.ai/blog/how-we-index-images-for-rag#HackerNews #Tech #AI
Structural retrieval shows promise over basic RAG for agent failure prediction on trajectory snapshots
A structural retrieval engine with KNN voting shows a promising signal over basic RAG: AUC = 0.705...
I made a personalized AI web app with RAG
So during my 4th Year of my college my team had decided to build a Personalized AI Assistant that can...
Permission-Aware Retrieval: The Missing Layer in Enterprise RAG Security
Most enterprise RAG systems are built as if retrieval is just search. That is the architectural...
Adding web search to our RAG pipeline: what broke and why
We added web search to a local RAG pipeline. It looked straightforward. It wasn’t. What...
Контекст — не инфраструктура: почему большое окно не заменяет retrieval-слойЗачем строить RAG или retrieval-слой, если с...
Контекст — не инфраструктура: почему большое окно не заменяет retrieval-слойЗачем строить RAG или retrieval-слой, если современные модели уже умеют работать с огромным контекстом?h...
دليل IBM الجديد "RAG Cookbook" يشرح كيف نبني أنظمة استرجاع معتمد على نماذج اللغة الكبيرة. 🔹 **مفتاح النجاح:** دمج محرك ب...
دليل IBM الجديد "RAG Cookbook" يشرح كيف نبني أنظمة استرجاع معتمد على نماذج اللغة الكبيرة. 🔹 **مفتاح النجاح:** دمج محرك بحث محلي مع نموذج LLM يرفع دقة النتائج ويقلل زمن الاستجابة، م...
RAG gives models memory.Training gives them their nature.And when context becomes incomplete, the scorpion returns.My la...
RAG gives models memory.Training gives them their nature.And when context becomes incomplete, the scorpion returns.My latest thoughts on local AI, Ruby, and why RAG isn't a magic s...