AI Intelligence Archive

Originally published at cchinchilla.dev. Part 2 of From code to weights, a 12-part series on ML...

Reinforcement learning has become a central component of large language model (LLM) post-training, yet token-level credit lacks a generally accepted mathematical definition, leaving its relationship to commonly used training signals unclea…

Quick answer: As of 4 days ago, September 18, Claude Code natively reads AGENTS.md, the open,...

Concept drift, driven by the rapid evolution of Android malware, severely degrades the performance of machine learning detectors. Current adaptation strategies are often reactive, responding only after performance has dropped and imposing …

The scarcity of non-English language data in specialized domains significantly limits the development of effective Natural Language Processing (NLP) tools. We present TransBERT, a novel framework for pre-training language models using excl…

Political discourse is widely perceived to be growing more hostile, yet robust evidence remains scarce. This study examines blame attribution in the Danish Parliament from 1997 to 2026, combining a purpose-built classifier, BlameBERT (F1: …

Residual Vector Quantization turns continuous representations into discrete, multi-level token sequences. Yet most methods operate in Euclidean space, despite the coarse-to-fine structure of the resulting codes and the latent hierarchies p…

Argument Mining (AM) transforms natural language into its underlying argument structures. This transformation is typically realized through a sequence of AM tasks that form an end-to-end AM pipeline. However, AM approaches often differ in …

Prefill and decode reward different approaches to quantization: low-precision arithmetic accelerates prompt processing, while compact weights reduce memory traffic during generation. We propose "disaggregated quantization" (DQ), which spec…

We study preference elicitation under the Bradley-Terry-Luce (BTL) model where the true partworth vector is unknown and has to be estimated as a parameter with elicited preference information. The set of selected pairwise queries is non-un…

TypeSafe launched Jev on September 15 and it reached OpenRouter three days later. Since then nearly...

Embodied world models predict the outcomes of robot actions to support learning and planning. For robots equipped with head and wrist cameras, this requires complementary views: the head view captures the overall task, while wrist views re…

Most posts about AI agents show you the demo. The demo always works. Nobody writes about the part...

The full matrix was 83 agents × 30 scenarios = 2,490 runs. Each one a real LLM call, 30–80 seconds....

I ship an Android app (Onira) that generates a personalized hypnosis/relaxation script on-device with...

Your LLM looks fine in the demo. Then it ships. Somewhere between "it works on my machine" and "a...

On July 22 I typed a one-line prompt on a test app: a widget built on a well-known REST API, the kind...

This blog turned one month old this week. I'd built a habit of checking four dashboards every few...

Every PR description is a claim. "No consumers of this API were changed." "Safe to merge." Most...

Most agent tasks do not need a frontier model. The subtasks inside a production agent are largely...

probe

Every time I needed to explore a dataset quickly, the same wall appeared: spin up a notebook, deal...

Old code, new AI, same unwritten test suite. TL;DR: Patch code that never had a test written for...

Retrieval-Augmented Generation (RAG) sounds complex, but the core concept is straightforward: instead...

I always had a slightly bad feeling about the credentials my coding agent could access. Did I secure...

A transaction hash tells us that something happened onchain. It does not tell us which information...

I built ChibiRigKit, an open-source tool that turns a single character illustration into an animated...

The European Commission has received the final report from its Special Panel on Child Safety Online,...

The agent stopped the training run. The GPU cluster did not. Agent Evaluation Case #003 ...

Most AI agent architectures use the same large language model for every task: planning, writing, tool...

Keyword Research with AI Agents and Google Search Console Manual keyword research is slow...

The record said there was a gas leak. It did not say where. { "urgency": "emergency", ...

What Did You Deliberately Leave Out of Your MVP? Most MVP posts show what made it into...

We write more tests than ever, yet bugs still reach production. This post introduces a risk-based...

Intro: Jev launched Sep 15 and drew a 140k waitlist in days. On Sep 20 TypeSafe removed the waitlist...

Let’s be honest: we all have a graveyard of bookmarks. Every time I see a cool open-source CLI tool...

There are developers who make a team faster. There are developers who make a team better. And there...

In a comment some weeks ago (link) i wrote about how it doesn't matter if you are using AI or not, or...

On September 5 I deleted 1,782 files from my autopilot's repository in one commit. Agents, templates,...

In this video I continue the fine-tuning series on my channel. This time I go through the whole...