Brian Cunningham

[hardmaru @hardmaru]

For over a decade, we’ve accepted that end-to-end backprop is the only way to train deep networks. But holding the entire network in memory all at once is why AI training is hitting a resource wall.

We found a new way to break the network into blocks and train them independently. The trick? Treating the network’s forward pass like a diffusion model denoising a signal.

This reinterpretation slashes the memory needed to train deep models. In our #ICLR2026 paper (https://t.co/PK5h0mqQSo), we matched end-to-end performance across ViTs, DiTs, and LLMs. We did this while training just one isolated block at a time.

[Brian Cunningham @TheRealBrianC]

@higgsri @jessesingal …and needn’t be discussed until the content of his confession is made public. From politicians to pundits to cops, everyone should have just checked themselves.

But Mark and Elon built platforms to enrich themselves off of the flames, and everyone felt obligated to pay up.