Experimental Intelligence Lab

Home About Research Data Documentation

Experimental
Intelligence Lab

We have a theory: Intelligence is too weird for one field to understand. So we experiment at the intersections — between every field that ever studied “thinking” and every machine that's trying to “think”.

Supported by

Research

View all

Classical landscape painting of Mount Pelion — home of the centaur Chiron — with a ruined temple and two horses above the Aegean

Conference

COLM 2026

Small Foundation Models of Human Cognition and Behaviour

Seventy billion parameters are not what cognitive foundation models rest on. Fourteen open models, 135M–14B, prove it.

A 17th century painting of Chickens and Ducks by Melchior d' Hondecoeter

Workshop

NeurIPS 2024

System 1.5

We gave machines the cognitive gear shift experts have — knowing when to think fast versus slow.

Cropped close up of dark green diagonal line on a pale green background

Workshop

NeurIPS 2025

Monitor-Generate-Verify

We translated metacognition into mathematics. AI now assesses strategy before attempting solutions.

Cropped close up of bright blue arc on a blue background

Workshop

COLM 2025

Before You 〈think〉, monitor

When AI pauses to assess difficulty first, it solves 7% more problems correctly.

Data

View all

Conference

ICWSM 2024

sentibank: Sixty years of human emotion dictionaries, unified

How we coded feelings before neural networks.

Conference

AIES 2025

PETLP: The legal stack for social media AI research, decoded

From extraction to model release.

Every AI lab builds forward. We build backward too. Maybe the solution to tomorrow's problem was solved in 1979. Or 1781.

Meet the lab

Documentation

View all

Cropped close up of light blue diagonal line on a bright green background

RedditHarbor

Reddit scraping with compliance built-in, not bolted on.

The 19th century painting, The Path By The Lake/ Ben Voirlich, by David Cox Junior

sentibank

Sixty years of emotion dictionaries. One import statement.

Socials

GitHub LinkedIn Instagram

Information

Home About Research Data Documentation

Website by

Thomas Bugg Anthony Odu

ExperimentalIntelligence Lab

We have a theory: Intelligence is too weird for one field to understand. So we experiment at the intersections — between every field that ever studied “thinking” and every machine that's trying to “think”.

Supported by

Research

Small Foundation Models of Human Cognition and Behaviour

System 1.5

Monitor-Generate-Verify

Before You 〈think〉, monitor

Data

sentibank: Sixty years of human emotion dictionaries, unified

PETLP: The legal stack for social media AI research, decoded

Every AI lab builds forward. We build backward too. Maybe the solution to tomorrow's problem was solved in 1979. Or 1781.

Documentation

RedditHarbor

sentibank

Experimental
Intelligence Lab