MMD (Maximum Mean Discrepancy) has shown promise in image generation. Can it improve diffusion language models too?
We post-train DLMs to match generated and reference distributions in a frozen DLM’s feature space. It works for both discrete and continuous models, and scales to 16B! 🚀
The results are strong across text generation, math reasoning, and code! Here’s how it works 👇
We post-train DLMs to match generated and reference distributions in a frozen DLM’s feature space. It works for both discrete and continuous models, and scales to 16B! 🚀
The results are strong across text generation, math reasoning, and code! Here’s how it works 👇
1:06
Introducing a simple method to post-train both discrete and continuous diffusion LMs.
It's based on Maximum Mean Discrepancy (MMD) that matches token-level feature distributions. Our method improves generation and speed on OWT/GSM8K and scales to 16B models. 🧵View quoted post on X
