[ICML 2026] Scaling Beyond Masked Diffusion Language Models
-
Updated
Sep 7, 2026 - Python
#
diffusion-llms
Here are 3 public repositories matching this topic...
This work introduces Flow Matching Mixture of Experts (FM-MoE), a framework that replaces conventional MLP experts with flow matching networks. Each expert learns a continuous transformation through an ordinary differential equation (ODE), enabling more expressive feature mappings while maintaining the computational efficiency of sparse experts
ai tensorflow transformers pytorch moe tensors diffusion-models flow-matching agents-sdk llm-space diffusion-llms
-
Updated
Feb 4, 2026 - Python
A curated reading list for awesome self-distillation works in LLMs, MLLMs, reasoning models, diffusion LLMs, and large generative models.
awesome reinforcement-learning awesome-list self-distillation llms mllms reasoning-models diffusion-llms large-generative-models
-
Updated
May 26, 2026
Add this topic to your repo
To associate your repository with the diffusion-llms topic, visit your repo's landing page and select "manage topics."