Good article about this topic

https://medium.com/@gedanken.thesis/the-loop-is-back-why-hrm-is-the-most-exciting-ai-architecture-in-years-7b8c4414c0b3

brain with two clocks

This temporal separation is what prevents the system from getting stuck on a bad idea.

it provides a revolutionary blueprint for how to assemble the powerful components we already have (Transformer blocks) into something far more capable for deep reasoning

It replaces the lone genius with a focused, two-person company — and as we’ll see, that makes all the difference.

Their collaboration unfolds in a beautifully nested loop of “thinking sessions,” “strategic meetings,” and “work sprints.”

Adaptive Computation Time

structured, disciplined, and multi-layered conversation between a strategist and an executor, repeated until the puzzle is solved

deep supervision

HRM’s genius is not in reinventing this engine, but in training two of them for specialized, cooperative roles.

The HRM model says: “We are only going to assign blame for the most recent action.”

The consequence of that buried mistake is no longer in the past; it’s the first item on the current agenda.

This is how HRM learns. It uses an incredibly efficient but myopic gradient, and brilliantly compensates for the myopia by using its deep, recursive loop to force itself to confront the consequences of its past mistakes, session after session, until the logic is perfected.

highly organized, adaptive reasoning engine excels at any task that requires deep, structured thought

It learns the deep structure of a problem with incredible data efficiency

HRM is a pure reasoning engine, designed to unravel this kind of deep, formal logic

Referenced in:

All notes