AI Research Posts

Guest Post: Optimizing Multi-agent Systems with Mistral Large, Mistral Nemo, and Llama-agents*

Agents can handle complex tasks with minimal human intervention. Learn how to build such agents with Mistral Large, Nemo, Llama agents, and Milvus.

FOD#65: Jevons' Paradox in AI

we discuss what the rapidly decreasing token cost of using LLMs means for AI companies, introduce some changes to Turing Post, and, as always, offer you the best-curated list of news and papers

10 Newest Ways for Efficient Processing of Long Context in LLMs

Inside DeepSeek Models

How DeepSeek-V2 and DeepSeek-Coder-V2 work: the DeepSeekMoE architecture, Multi-Head Latent Attention (MLA), and what makes them efficient

FOD#64: Golden Age for Indie Devs and Engineers

How Cursor AI and LeRobot are Empowering a New Generation of Developers to Innovate and Compete.

Mamba Architecture Research: 15 Papers With Open Code

15 open-source Mamba architecture papers: Jamba-1.5, MambaVision, BioMamba, Speech Slytherin, and more. Covering text, vision, audio, and hybrid models.

FOD#63: Open-Ended Exploration

or how we become AI-augmented PLUS the news from the usual suspects, and the hands-down best-curated list of last week's papers and models

What is Speculative RAG

Speculative RAG uses a small drafter and a large verifier LM to boost RAG speed and accuracy. How it works, where it excels, and key limitations.

Guest Post: How to Use Function Calling with Ollama, Llama3 and Milvus*

A step-by-step guide on how to integrate Llama 3.1 with external tools like Milvus vector database and APIs to build powerful, context-aware applications

FOD#62: DeepMind’s New Techniques Are Shaping the Future – Here’s How

Plus, a new rubric on AGI, Karpathy's take on RL, news from the usual suspects, and the hands-down best-curated list of last week's papers and models

12 Types of Retrieval-Augmented Generation (RAG)

12 RAG techniques explained: GraphRAG, Self-RAG, LongRAG, Modular RAG& more. Find the right retrieval-augmented generation approach for your LLM