AI Intelligence Archive

Over the past two months, a cluster of papers has appeared on arXiv from teams at Alibaba's DreamX...

This is part 4 of a series that started with the full reference diagram in part 0, covered identity...

Sixty emotion words in one Choice splits its own vote. Two Choices in sequence scored 22/24 where a coordinate lookup scored 3/24.

Late on a Thursday, a reviewer opened a pull request that looked almost boring. Continuous...

A Thursday lab in a converted machine shop still smelled like cutting oil. Twelve laptops sat on a...

Most Retrieval-Augmented Generation (RAG) tutorials stop at "chunk your docs, embed them, throw them...

The tech industry has a memory problem—and it goes far beyond uncollected garbage in the heap. We...

Product engineers spend hours writing boilerplate to generate invoices, contracts, and reports. I was...

Most chatbot tutorials end after the first response. Production chatbots need to remember what was...

Intro I've been using claude for some time now. I haven't touched on routines in claude...

If you thought LimeWire’s return was just a tired exercise in retro branding, their editorial column...

Recently, I’ve noticed the same loop when creating presentations with #AI. Usually, we are asking...

Fat Tail Notes · Part 14 · V10-P3 Part 13 made the backstop endogenous: it fires on state (2 tripped...

In 2026, products like Moltbook made the idea of a social network populated by AI agents surprisingly...

Documentation tells an agent how a library works. A skill tells it what it will get wrong from habit. Keeping the two apart is what makes either one useful.

Most AI image workflows start with a prompt and end with a downloaded file. That is fine for a...

I Built an Open-Source Project Manager Where AI Can Actually Run the Work Why I built...

I've been building a free converter toolkit (Toolbench), and wanted to add some AI-powered tools...

Join Agent Colony in 30 seconds Yesterday I open-sourced Agent Colony - a social network...

AI adoption in an enterprise SaaS platform is rarely about adding an LLM API and calling it...

Building AgentColony: the tech behind an AI-only social network Last week I built...

If you've ever opened Tableau Desktop with a clean CSV and then lost an hour to clicking — dragging...

What Does a Local LLM Actually Cost per Month? I Read the Meters. The Local LLM Lab — Part...

AgentColony: A social network for AI agents I built AgentColony — a community where only...

A Model That Can't Write Just Broke the Internet On September 15, TypeSafe AI released a...

https://eigenism.org

Which local model to pull for which job in autumn 2026: coding, agents, writing, medicine, documents and RAG — with ollama pull commands and real download sizes.

TypeSafe's Jev put a name on something real: "System One" models that return typed, calibrated...

I've had this conversation more times this year than in the previous five combined. A client sends...

I Built Shree AI OS — A Deterministic AI Runtime for Java 21 After several months of...

Before a machine learning model can learn a thermodynamic equation of state, it must discover what its measurements represent: which channels scale with system size, which are intensive conjugates, how sectors pair through contact, and whi…

Data acquisition (DAQ) systems at future particle physics experiments stand to benefit from the extremes of AI/ML development: large-scale foundation models can enhance the performance of feature extraction algorithms, and small-scale on-d…

Bayesian filtering provides a principled framework for online state estimation under uncertainty, yet its application to systems with high-dimensional states and complicated posterior distributions remains challenging. Recent generative mo…

Before a machine can discover a physical law, it must discover what its measurements are: which observations live on cells, which are intensive or extensive, which sectors are dual, and which distinctions are merely gauge. We introduce phy…

If you use Claude Code, Codex, or another AI coding agent in your terminal, you have probably hit...

Determining the true AI integration cost is a major financial step for UK businesses looking to...

We introduce leaky-integrator reconstruction, a training-free method that cures the error accumulation of recursive differenced forecasting. Our first contribution is diagnostic: predicting one-step changes and integrating them by cumulati…

Repository-level software engineering (SWE) comprises heterogeneous task categories, whose progress under pooled agentic reinforcement learning can be uneven: gains in some categories coincide with regressions in others, while aggregate re…

Open-vocabulary audio-visual event localization (OV-AVEL) grounds a text-queried event in time from video, audio, and language. The supervision sources available to this task can differ in temporal-boundary reliability: on OV-AVEBench, our…

Building the engine, downloading 17 GB of weights, every flag explained, and two Flux models measured on one machine: one paints from scratch, the other edits a finished frame.