---
library_name: transformers
license: apache-2.0
datasets:
- billion-word-benchmark/lm1b
---

## Quick Start Guide

To use this pre-trained model with the HuggingFace APIs, use the following snippet:

```python
from transformers import AutoModelForMaskedLM, AutoTokenizer

# See the `UDLM` collection page on the hub for list of available models.
tokenizer = transformers.AutoTokenizer.from_pretrained('bert-base-uncased')
model_name = 'kuleshov-group/udlm-lm1b'
model = AutoModelForMaskedLM.from_pretrained(model_name)
```


## Model Details

UDLM stands for **U**niform **D**iffusion **L**anguage **M**odels.
This model was trained using the refined uniform noise discrete diffusion continuous-time ELBO introduced [here](https://arxiv.org/abs/2412.10193).

### Architecture

The model has a context size of 128 tokens. The model has 139M parameters.

The model architecture is based off of the [Diffusion Transformer architecture](https://arxiv.org/abs/2212.09748) and consists of:
- 12 multi-head attention blocks (with 12 attention heads),
- hidden dimension of 768,
- `adaLN` for conditioning on time-step (i.e., during diffusion training / generation).


### Training Details

The model was trained using the `bert-base-uncased` tokenizer.
We trained for 1M gradient update steps using a batch size of 512.
We use linear warm-up with 2500 steps until we reach a constant learning rate of 3e-4.

For more details, please refer to our work: [Simple Guidance Mechanisms for Discrete Diffusion Models](https://arxiv.org/abs/2412.10193).

## Citation
Please cite our work using the bibtex below:

### BibTeX:
```
@article{schiff2024discreteguidance,
  title={Simple Guidance Mechanisms for Discrete Diffusion Models},          
  author={Schiff, Yair and Sahoo, Subham Sekhar and Phung, Hao and Wang, Guanghan and Boshar, Sam and Dalla-torre, Hugo and de Almeida, Bernardo P and Rush, Alexander and Pierrot, Thomas and Kuleshov, Volodymyr},
  journal={arXiv preprint arXiv:2412.10193},
  year={2024}
}
```