---
license: apache-2.0
base_model:
- meta-llama/Llama-3.1-8B-Instruct
language:
- zh
- en
---
# CSLM-base

## 💡 Model Description

`CSLM-base` is a base cross-lingual speech language model checkpoint from the ACL 2026 Findings project *Efficient Training for Cross-lingual Speech Language Models*.  
It is designed for speech-to-speech / speech-conditioned generation workflows where speech units are used as model input.

Paper: https://arxiv.org/abs/2604.11096

## 📖 Citation

If you use this model, please cite:

```bibtex
@misc{zhou2026efficienttrainingcrosslingualspeech,
      title={Efficient Training for Cross-lingual Speech Language Models}, 
      author={Yan Zhou and Qingkai Fang and Yun Hong and Yang Feng},
      year={2026},
      eprint={2604.11096},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2604.11096}, 
}
```

## ✉️ Contact

For questions, contact: `zhouyan23z@ict.ac.cn`