AI Intelligence Archive

What if you could have an AI coding agent that reads your files, writes code, executes commands, and...

Every restart erased our agents' memory. Every naive relaunch made them redo work they had already done, emails included. Here's how we made them remember everything and relaunch only when it's safe.

Multilingual dense retrievers place all languages in one shared space so that a question finds relevant documents, regardless of which language they are written in. We show that the same question, asked in different languages, retrieves su…

Task planning for robots in off-site construction prefabrication is primarily manual, leading to planning inefficiencies and reluctance to adopt robotic solutions. Existing approaches provide limited support for scalable, modular automated…

Reusable skills let LLM agents load task-specific instructions on demand, but they make execution depend on implicit routing over skill metadata. We test whether this design scales by holding SkillsBench tasks and their golden skills fixed…

Autoregressive language models produce, at every prediction step, a vocabulary-sized logit vector whose structure determines the conditional next-token distribution. We empirically identify that the ordered vocabulary-head logits follow a …

Knowledge distillation trains student models on teacher-generated responses to transfer task capability, but those responses may also carry behavioral tendencies of their source. We investigate whether the source teacher can shape student …

Post-training pruning enables efficient deployment, but conventional methods do not explicitly preserve internal-gradient alignment between sparse and dense models. We study novel connections between downstream performance of compressed mo…

Large language models (LLMs) increasingly act as personal assistants; however, little is known about how they capture national-level preferences on a global scale. Given that "time" is a universal and representative concept, we examine LLM…

Motivated by applications to cognitive radio networks, multi-player multi-armed bandits (MMAB) have attracted considerable attention in recent years. The asynchronous variant, in which players are active only at a fraction of timesteps, be…

Buildings cause nearly 40\% of carbon emissions in the US. Of total building emissions, HVAC systems are responsible for approximately 40-60\% and account for 15\% of global energy consumption. Despite advances in artificial intelligence, …

The safety of LLM-based agents currently rests on alignment training and safety classifiers, neither of which can rule out that a sophisticated indirect prompt injection (IPI) goes through and causes significant harm. Recent work therefore…

Clinical trial protocols serve as the source of truth for study design, but are mostly written and used as unstructured PDF documents. Digitizing protocols turns static content into structured, machine-readable data that clinical software …

Recently, Morris et al. introduced TinyLoRA, a strategy for LLM fine-tuning which introduces arbitrarily few parameters, and demonstrated accuracy improvements using less than $10$ parameters rather than the ${\gtrsim 10,000}$ typically de…

AI is everywhere. I don't know how to start this off really haha. Though to be fair, everyone is...

Egocentric video understanding spans diverse capabilities, modalities, and evaluation formats, yet these are often studied in isolation despite their tight coupling in real-world interaction. We propose the Ego-Action-World interaction loo…

Autoformalization (AF) requires more than compilation: formal statements must preserve the meaning of their informal sources, a property known as faithfulness. An emerging frontier is the automated evaluation of this property, yet most exi…

Tool-call judges evaluate argument values proposed by LLM agents. We find that they accept a correct value less often when their input holds no reference for it, that is, no text that says what the value should be. This happens when the va…

In order to execute quantum algorithms on bosonic quantum computers, it is necessary to encode their inputs into a desired target quantum state, with a process called state preparation. In real settings, state preparation methods operate i…

Objective digital measures of depression severity could complement conventional symptom assessments and facilitate more timely clinical evaluation. Here, we investigated whether task-evoked eye movements provide interpretable behavioral ma…

Many related prediction problems, which we call questions, come with few labels each. What information can be shared across them, and what is actually necessary for good decisions? We answer both for recalibration, motivated by decision mo…

Membership Inference Attacks (MIAs) are currently the dominant approach to evaluating privacy in machine learning applications. Despite their significance, several concerns remain unexplored, particularly with regard to tabular data. Exist…

Terminal agents have shown strong abilities in complex coding tasks. However, they often struggle to recover from errors during long-running execution (e.g., command errors, faulty implementations), eventually leading to task failures. To …

A while ago I forked a running agent session three ways and wrote about it. Three people in the...

A beginner's reflection on the kLab TekHer AI Program — Cohort 2 When I started the "Introduction to...

Claiming an LLM agent's memory poisoned its behaviour is easy. Proving it is not. A measurement rig built on counterfactual amputation, a noise floor, and three honest negative results.

Here's a failure your test suite probably can't see. An AI agent with a send_email tool summarizes a...

Every era of the cloud has a unit of work. For agents, it’s a human who directs, a controller who...

I spent a weekend running two small language models against three jobs I would normally hand to a...

What I learned on Day 1 of the kLab TekHer AI Program, Cohort 2 — how images are just numbers, how filters reveal edges, how CNNs learn, and why computer vision matters for Rwanda and Africa.

On October 13, 2026, banking regulators intensified scrutiny of autonomous AI agents. The...

A weekend AWS agent that inspects any public GitHub repo, scores it for safety and trust, flags risky installs like curl-pipe-bash, and guides you to a working first run. Here is how I built it on Lambda and Bedrock.

Why I wanted this Grok Bot runs a lot of my day. I don't have one assistant, I have a team...

You can’t vibe code your way to a great customer experience. You need the user flows and the user...

I've been building Kervan for a while. It turns HTTP APIs into tools AI assistants can use. You...