Publications

2026

1. arXiv

   Muon^2: Boosting Muon via Adaptive Second-Moment Preconditioning

   Ziyue Liu, Ruijie Zhang, Zhengyang Wang, and 4 more authors

   arXiv preprint arXiv:2604.09967, 2026

   Bib HTML

   

   @article{liu2026muon2,
     author = {Liu, Ziyue and Zhang, Ruijie and Wang, Zhengyang and Zhao, Yequan and Su, Yupeng and Yang, Zi and Zhang, Zheng},
     journal = {arXiv preprint arXiv:2604.09967},
     title = {Muon$^2$: Boosting Muon via Adaptive Second-Moment Preconditioning},
     year = {2026},
   }

2. arXiv

   Muon+: Towards better muon via one additional normalization step

   Ruijie Zhang, Yequan Zhao, Ziyue Liu, and 2 more authors

   arXiv e-prints, 2026

   Bib HTML

   

   @article{zhang2026muonplus,
     author = {Zhang, Ruijie and Zhao, Yequan and Liu, Ziyue and Wang, Zhengyang and Zhang, Zheng},
     journal = {arXiv e-prints},
     title = {Muon+: Towards better muon via one additional normalization step},
     year = {2026},
     pages = {arXiv: 2602.21545},
   }

3. arXiv

   TEON: Tensorized Orthonormalization Beyond Layer-Wise Muon for Large Language Model Pre-Training

   Ruijie Zhang, Yequan Zhao, Ziyue Liu, and 5 more authors

   arXiv preprint arXiv:2601.23261, 2026

   Bib HTML

   

   @article{zhang2026teon,
     author = {Zhang, Ruijie and Zhao, Yequan and Liu, Ziyue and Wang, Zhengyang and Li, Dongyang and Su, Yupeng and Liu, Sijia and Zhang, Zheng},
     journal = {arXiv preprint arXiv:2601.23261},
     title = {TEON: Tensorized Orthonormalization Beyond Layer-Wise Muon for Large Language Model Pre-Training},
     year = {2026},
   }

4. arXiv

   ReCoVer: Resilient LLM Pre-Training System via Fault-Tolerant Collective and Versatile Workload

   Ziyue Liu, Zhengyang Wang, Ruijie Zhang, and 7 more authors

   arXiv preprint arXiv:2605.11215, 2026

   Bib HTML

   

   @article{liu2026recover,
     author = {Liu, Ziyue and Wang, Zhengyang and Zhang, Ruijie and Maurya, Avinash and Zhou, Hui and Hovland, Paul and Di, Sheng and Cappello, Franck and Nicolae, Bogdan and Zhang, Zheng},
     journal = {arXiv preprint arXiv:2605.11215},
     title = {ReCoVer: Resilient LLM Pre-Training System via Fault-Tolerant Collective and Versatile Workload},
     year = {2026},
   }

2025

1. EMNLP

   Cola: Compute-efficient pre-training of llms via low-rank activation

   Ziyue Liu^*, Ruijie Zhang^*, Zhengyang Wang^*, and 5 more authors

   In EMNLP 2025 (oral), 2025

   Bib HTML

   

   @inproceedings{liu2025cola,
     author = {Liu, Ziyue and Zhang, Ruijie and Wang, Zhengyang and Yang, Zi and Hovland, Paul and Nicolae, Bogdan and Cappello, Franck and Zhang, Zheng},
     booktitle = {EMNLP 2025 (oral)},
     title = {Cola: Compute-efficient pre-training of llms via low-rank activation},
     year = {2025},
   }

2. NeurIPS

   LaX: Boosting Low-Rank Training of Foundation Models via Latent Crossing

   Ruijie Zhang, Ziyue Liu, Zhengyang Wang, and 1 more author

   In NeurIPS 2025, 2025

   Bib HTML

   

   @inproceedings{zhang2025lax,
     author = {Zhang, Ruijie and Liu, Ziyue and Wang, Zhengyang and Zhang, Zheng},
     booktitle = {NeurIPS 2025},
     title = {LaX: Boosting Low-Rank Training of Foundation Models via Latent Crossing},
     year = {2025},
   }

3. MLSys

   BOOST: BOttleneck-Optimized Scalable Training Framework for Low-Rank Large Language Models

   Zhengyang Wang^*, Ziyue Liu^*, Ruijie Zhang, and 5 more authors

   In MLSys 2026, 2025

   Bib HTML

   

   @inproceedings{wang2025boost,
     author = {Wang, Zhengyang and Liu, Ziyue and Zhang, Ruijie and Maurya, Avinash and Hovland, Paul and Nicolae, Bogdan and Cappello, Franck and Zhang, Zheng},
     booktitle = {MLSys 2026},
     title = {BOOST: BOttleneck-Optimized Scalable Training Framework for Low-Rank Large Language Models},
     year = {2025},
   }

4. MLSys

   SkipKV: Selective Skipping of KV Generation and Storage for Efficient Inference with Large Reasoning Models

   Jiayi Tian, Seyedarmin Azizi, Yequan Zhao, and 7 more authors

   In MLSys 2026, 2025

   Bib HTML

   

   @inproceedings{tian2025skipkv,
     author = {Tian, Jiayi and Azizi, Seyedarmin and Zhao, Yequan and Potraghloo, Erfan Baghaei and McPherson, Sean and Sridhar, Sharath Nittur and Wang, Zhengyang and Zhang, Zheng and Pedram, Massoud and Kundu, Souvik},
     booktitle = {MLSys 2026},
     title = {SkipKV: Selective Skipping of KV Generation and Storage for Efficient Inference with Large Reasoning Models},
     year = {2025},
   }

2023

1. MICRO

   Rm-stc: Row-merge dataflow inspired gpu sparse tensor core for energy-efficient sparse acceleration

   Guyue Huang, Zhengyang Wang, Po-An Tsai, and 3 more authors

   In MICRO 2023. Proceedings of the 56th Annual IEEE/ACM International Symposium on Microarchitecture, Oct 2023

   Bib HTML

   

   @inproceedings{huang2023rmstc,
     author = {Huang, Guyue and Wang, Zhengyang and Tsai, Po-An and Zhang, Chen and Ding, Yufei and Xie, Yuan},
     booktitle = {MICRO 2023. Proceedings of the 56th Annual IEEE/ACM International Symposium on Microarchitecture},
     title = {Rm-stc: Row-merge dataflow inspired gpu sparse tensor core for energy-efficient sparse acceleration},
     year = {2023},
     month = oct,
     pages = {338--352},
   }