metadata
license: mit
pipeline_tag: text-generation
ELF: Embedded Language Flows
ELF is a class of continuous diffusion language models based on continuous-time Flow Matching, presented in the paper ELF: Embedded Language Flows.
Unlike existing diffusion language models (DLMs), ELF predominantly stays within the continuous embedding space until the final time step, where it maps to discrete tokens using a shared-weight network. This formulation makes it straightforward to adapt established techniques from image-domain diffusion models, such as classifier-free guidance (CFG).
- Paper: ELF: Embedded Language Flows
- Repository: https://github.com/lillian039/ELF
Inference and Evaluation
This checkpoint can be used with the official JAX implementation. To verify the setup or run evaluation, follow these steps:
Initialization
git clone https://github.com/lillian039/ELF
cd ELF
pip install -r requirements.txt
Running Evaluation
To evaluate the model on unconditional generation (OpenWebText):
cd src/
# For ELF-B (105M)
python eval.py \
--config configs/training_configs/train_owt_ELF-B.yml \
--checkpoint_path embedded-language-flows/ELF-B-owt
To evaluate on conditional tasks like summarization (XSum):
cd src/
python eval.py \
--config configs/training_configs/train_xsum_ELF-B.yml \
--checkpoint_path embedded-language-flows/ELF-B-xsum
Citation
@article{elf2026,
title={ELF: Embedded Language Flows},
author={Hu, Keya and Qiu, Linlu and Lu, Yiyang and Zhao, Hanhong and Li, Tianhong and Kim, Yoon and Andreas, Jacob and He, Kaiming},
journal={arXiv preprint arXiv:2605.10938},
year={2026}
}