Leo Zhang
@leoeleoleo1.bsky.social
PhD student at @oxfordstatistics.bsky.social working on generative models
Pinned
Leo Zhang
@leoeleoleo1.bsky.social
· Mar 4
In our new paper (accepted at ICLR), we propose the first framework for constructing equivariant diffusion models via symmetrisation
This allows us to ensure E(3)-equivariance with just highly scalable standard architectures such as Diffusion Transformers, instead of EGNNs, for molecular generation
This allows us to ensure E(3)-equivariance with just highly scalable standard architectures such as Diffusion Transformers, instead of EGNNs, for molecular generation
Reposted by Leo Zhang
A meta-point of this paper is that category theory has utility for reasoning about current problems of interest in mainstream machine learning. The theory is predictive, not just descriptive. 🧵(1/6)
In our new paper (accepted at ICLR), we propose the first framework for constructing equivariant diffusion models via symmetrisation
This allows us to ensure E(3)-equivariance with just highly scalable standard architectures such as Diffusion Transformers, instead of EGNNs, for molecular generation
This allows us to ensure E(3)-equivariance with just highly scalable standard architectures such as Diffusion Transformers, instead of EGNNs, for molecular generation
March 6, 2025 at 4:38 AM
A meta-point of this paper is that category theory has utility for reasoning about current problems of interest in mainstream machine learning. The theory is predictive, not just descriptive. 🧵(1/6)
In our new paper (accepted at ICLR), we propose the first framework for constructing equivariant diffusion models via symmetrisation
This allows us to ensure E(3)-equivariance with just highly scalable standard architectures such as Diffusion Transformers, instead of EGNNs, for molecular generation
This allows us to ensure E(3)-equivariance with just highly scalable standard architectures such as Diffusion Transformers, instead of EGNNs, for molecular generation
March 4, 2025 at 3:31 PM
In our new paper (accepted at ICLR), we propose the first framework for constructing equivariant diffusion models via symmetrisation
This allows us to ensure E(3)-equivariance with just highly scalable standard architectures such as Diffusion Transformers, instead of EGNNs, for molecular generation
This allows us to ensure E(3)-equivariance with just highly scalable standard architectures such as Diffusion Transformers, instead of EGNNs, for molecular generation