Papers6

#masked diffusion

Balancing Understanding and Generation in Discrete Diffusion Models

This paper introduces XDLM, a single model that blends two popular diffusion styles (masked and uniform) so it both understands and generates text and images well.

#XDLM#discrete diffusion#stationary noise kernel

The Flexibility Trap: Why Arbitrary Order Limits Reasoning Potential in Diffusion Language Models

Beginner

Zanlin Ni, Shenzhi Wang et al.Jan 21arXiv

Diffusion language models can write tokens in any order, but that freedom can accidentally hurt their ability to reason well.

#diffusion language model#arbitrary order generation#autoregressive training

On the Role of Discreteness in Diffusion LLMs

Intermediate

Ziqi Jin, Bin Wang et al.Dec 27arXiv

The paper asks what a truly good diffusion-based language model should look like and lists five must-have properties.

#diffusion language models#smooth corruption#discrete tokens

Scaling Behavior of Discrete Diffusion Language Models

Intermediate

Dimitri von Rütte, Janis Fluri et al.Dec 11arXiv

This paper studies how a newer kind of language model, called a discrete diffusion language model (DLM), gets better as we give it more data, bigger models, and more compute.

#discrete diffusion#language models#scaling laws

LLaDA2.0: Scaling Up Diffusion Language Models to 100B

Intermediate

Tiwei Bie, Maosong Cao et al.Dec 10arXiv

Before this work, most big language models talked one word at a time (autoregressive), which made them slow and hard to parallelize.

#diffusion language model#masked diffusion#block diffusion

Learning Unmasking Policies for Diffusion Language Models

Intermediate

Metod Jazbec, Theo X. Olausson et al.Dec 9arXiv

Diffusion language models write by gradually unmasking hidden words, so deciding which blanks to reveal next is a big deal for both speed and accuracy.

#diffusion language models#masked diffusion#unmasking policy