Tomer Barak

Research Blog

Thoughts on automated science and AI research

Blog

Exploring AI-led science and the future of research

April 29, 2025

The mathematical foundations of scale theory: Bridging micro and macro phenomena

Scale Theory Multi-scale Systems Complex Systems

In our quest to understand the universe, we’ve developed sophisticated tools for investigating phenomena at particular scales—from quantum mechanics at the subatomic level to general...

Read More

April 27, 2025

Giving LLM agents a life of their own through self-messaging

BarakBot Self-Referential LLM Agents

BarakBot: agents that talk to themselves (and the world)

Read More

March 20, 2025

One more Copernican revolution: reconsidering scale-centrism

Scale Relativity Cosmology Metaphysics

The Copernican Revolution famously displaced Earth from the center of the cosmos, forcing humanity to confront its non-privileged position in space. However, I propose that...

Read More

March 07, 2025

Self-referential agent architecture: how BarakBot agents talk to themselves

BarakBot Agent Architecture LLM Agents

In developing BarakBot, I encountered a fundamental challenge in multi-agent coordination. How should distinct AI agents communicate effectively while keeping the architecture simple and scalable?...

Read More

March 06, 2025

Abstract reasoning without prior training: untrained neural networks solve visual intelligence tests

Abstract Reasoning Neural Networks Fluid Intelligence

In our recent paper published in Scientific Reports, we uncover the surprising ability of untrained artificial neural networks (ANNs) to solve abstract visual reasoning tasks....

Read More