kuleshov-group/mdlm

[NeurIPS 2024] Simple and Effective Masked Diffusion Language Model

49
/ 100
Emerging

This project offers a sophisticated tool for generating high-quality text by filling in masked or missing parts of sentences. It takes partial or garbled text as input and efficiently produces complete, coherent sentences or longer sequences. This is ideal for researchers and practitioners working on advanced language generation tasks.

657 stars. No commits in the last 6 months.

Use this if you need to generate high-quality, coherent text sequences quickly and efficiently, especially from incomplete or noisy inputs.

Not ideal if your primary need is for a simple, off-the-shelf text generation tool without needing to engage with underlying model architectures or advanced sampling techniques.

natural-language-generation text-synthesis language-modeling computational-linguistics
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 21 / 25

How are scores calculated?

Stars

657

Forks

91

Language

Python

License

Apache-2.0

Last pushed

Sep 29, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/kuleshov-group/mdlm"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.