AI Research Posts

🌁#79: Sora and World Models – Bringing magic to muggles

Spatial Intelligence just got a boost! Plus, a concise coverage of the remarkably rich week in ML research and innovations

16 New Types of Retrieval-Augmented Generation (RAG)

🦸🏻#6: The Role of Profiling in Agentic Workflows

How AI agents perceive, decide, and act: agent profiling explained through the BDI model, PEAS framework, performance benchmarks, and resource monitoring.

Guest post: When Search Meets AI: The Hybrid Revolution*

Inside Milvus Vector Database's Intelligent Search Solution

Flow Matching for Generative Modeling: How It Works and Why It Matters

Flow Matching explained: how it trains generative models faster than diffusion, what conditional flow matching adds, and why Flux, F5-TTS and MovieGen use it.

🌁#78: Enabling the Future of AI (2025)

join the prediction game plus our usual collection of interesting articles, relevant news, and research papers. Dive in!

Baichuan Intelligence: The AI Tiger Focused on Math and Healthcare

Inside Baichuan Intelligence — Wang Xiaochuan, the retreat from general-purpose AI, Baichuan-M1 through M4, and the high-risk bet on building an AI doctor.

LLaVA-o1: Step-by-Step Visual Reasoning VLM Explained

LLaVA-o1 reasons step-by-step through 4 structured stages & stage-level beam search. How it works, benchmarks vs GPT-4o-mini & Gemini-1.5-pro, and where it falls short.

🌁#77: Amid Big Model Chaos: Small Models and Embeddings Steal the Spotlight

plus our usual collection of interesting articles, relevant news, and research papers. Dive in!

12 Researches on Sparse Autoencoders

12 research papers on sparse autoencoders (SAEs): interpretability, LLM steering, feature extraction, and alignment.

🦸🏻#5: Building Blocks of Agentic Systems

What makes an AI agent work? A deep dive into profiling, knowledge,memory, reasoning, reflection & actions — the core components of agentic systems.

How to improve the observability and reliability of your AI cluster [💡Webinar ]