Blog
Featured
Video World Models
Video diffusion models can generate stunning short clips, but building true world simulators requires a fundamental shift. This post explores...
Hamiltonian Mechanics
Hamiltonian mechanics is a way to describe how physical systems, like planets or pendulums, move over time, focusing on energy...
All Posts
Video World Models
Video diffusion models can generate stunning short clips, but building true world simulators requires a fundamental shift. This post explores how recent advances in causal attention, distribution matching distillation, and...
Training an EBM with Contrastive Divergence
Interactive notebook providing a hands-on tutorial on training a simple EBM on synthetic datasets using the TorchEBM library. Covers energy functions, Langevin Dynamics for sampling, and Contrastive Divergence.
Hamiltonian Mechanics
Hamiltonian mechanics is a way to describe how physical systems, like planets or pendulums, move over time, focusing on energy rather than just forces. By reframing complex dynamics through energy lenses, this 19th-century physics framework now powers cutting-edge generative AI.
The Map Of Transformers
A broad overview of Transformers research covering the landscape of transformer architectures, their applications, and the evolution of attention mechanisms in deep learning.
Transformers in Action: Attention Is All You Need
A brief survey, illustration, and implementation guide to understanding the transformer architecture and attention mechanisms that revolutionized natural language processing.
Rethinking Thinking: How Do Attention Mechanisms Actually Work?
The brain, the mathematics, and DL – research frontiers exploring the intersection of neuroscience and attention mechanisms in artificial intelligence.