| --- |
| base_model: Qwen/Qwen3-8B-Base |
| library_name: ladir |
| pipeline_tag: text-generation |
| tags: |
| - code |
| - latent-reasoning |
| - diffusion |
| - flow-matching |
| --- |
| |
| # LaDiR-Code-8B |
|
|
| **LaDiR** (Latent Diffusion Reasoner) lets an LLM reason in a continuous latent space: a VAE compresses a reasoning |
| chain into 64 latent tokens, the LLM generates this latent block with flow matching, then decodes the answer |
| conditioned on it. This checkpoint is built on Qwen3-8B-Base for code generation. |
|
|
| Code: [github.com/mk322/LaDiR](https://github.com/mk322/LaDiR) · Paper: [arXiv:2510.04573](https://arxiv.org/abs/2510.04573) |
|
|
| ```bash |
| git clone https://github.com/mk322/LaDiR && cd LaDiR && pip install -e . |
| python -m eval.generate --checkpoint mk322/LaDiR-Code-8B |
| ``` |
|
|
| ```bibtex |
| @inproceedings{kang2026ladir, |
| title = {LaDiR: Latent Diffusion Enhances LLMs for Text Reasoning}, |
| author = {Kang, Haoqiang and Zhang, Yizhe and Kuang, Nikki Lijing and Majamaki, Nicklas and Jaitly, Navdeep and Ma, Yi-An and Qin, Lianhui}, |
| booktitle = {International Conference on Learning Representations (ICLR)}, |
| year = {2026}, |
| url = {https://arxiv.org/abs/2510.04573} |
| } |
| ``` |
|
|
