Zhixiong Zhao

^† Equal Contribution, ^* Corresponding Author(s)

1. ICML 2026

   TWLA: Breaking the Barrier to W1.58A4 Post-Training Quantization for LLMs

   Zhixiong Zhao^† , Zukang Xu^† , Zhixuan Chen , Xing Hu , Zhe Jiang , and Dawei Yang^*

   In International Conference on Machine Learning, 2026 (Top Conf. in AI) CCF-A

   Paper Code Cite

2. ACL-Main 2026

   BWLA: Breaking the Barrier of W1AX Post-Training Quantization for LLMs

   Zhixiong Zhao^† , Zukang Xu^† , and Dawei Yang^*

   In Annual Meeting of the Association for Computational Linguistics, 2026 (Top Conf. in NLP) CCF-A

   Paper Code Cite

   @misc{zhao2026bwlabreakingbarrierw1ax,
       title={BWLA: Breaking the Barrier of W1AX Post-Training Quantization for LLMs}, 
       author={Zhixiong Zhao and Zukang Xu and Dawei Yang},
       year={2026},
       eprint={2605.00422},
       archivePrefix={arXiv},
       primaryClass={cs.LG},
       url={https://arxiv.org/abs/2605.00422}, 
   }

3. ICLR 2026

   KBVQ-MoE: KLT-guided SVD with Bias-Corrected Vector Quantization for MoE Large Language Models

   Zukang Xu^† , Zhixiong Zhao^† , Xing Hu , Zhixuan Chen , and Dawei Yang^*

   In International Conference on Learning Representations, 2026 (Top Conf. in AI) CCF-A

   Paper Code Cite

   @article{xu2026kbvq,
     title={KBVQ-MoE: KLT-guided SVD with Bias-Corrected Vector Quantization for MoE Large Language Models},
     author={Xu, Zukang and Zhao, Zhixiong and Hu, Xing and Chen, Zhixuan and Yang, Dawei},
     journal={arXiv preprint arXiv:2602.11184},
     year={2026}
   }

4. AAAI 2026

   SpecQuant: Spectral Decomposition and Adaptive Truncation for Ultra-Low-Bit LLMs Quantization

   Zhixiong Zhao^† , Fangxin Liu^† , Junjie Wang , Chenyang Guan , Zongwu Wang , Li Jiang , and Haibing Guan^*

   In Association for the Advancement of Artificial Intelligence, 2026 (Top Conf. in AI) CCF-A

   Paper Code Cite

   @inproceedings{zhao2026specquant,
     title={Specquant: Spectral decomposition and adaptive truncation for ultra-low-bit llms quantization},
     author={Zhao, Zhixiong and Liu, Fangxin and Wang, Junjie and Guan, Chenyang and Wang, Zongwu and Jiang, Li and Guan, Haibing},
     booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
     volume={40},
     number={34},
     pages={28786--28794},
     year={2026}
   }

5. ICCAD 2025

   QUARK: Quantization-Enabled Circuit Sharing for Transformer Acceleration by Exploiting Common Patterns in Nonlinear Operations

   Zhixiong Zhao^† , Haomin Li^† , Fangxin Liu , Yuncheng Lu , Zongwu Wang , Tao Yang , Li Jiang , and Haibing Guan^*

   In International Conference on Computer-Aided Design, 2025 (Top Conf. in EDA) CCF-B

   Paper Code Cite

   @inproceedings{zhao2025quark,
     title={QUARK: Quantization-Enabled Circuit Sharing for Transformer Acceleration by Exploiting Common Patterns in Nonlinear Operations},
     author={Zhao, Zhixiong and Li, Haomin and Liu, Fangxin and Lu, Yuncheng and Wang, Zongwu and Yang, Tao and Jiang, Li and Guan, Haibing},
     booktitle={2025 IEEE/ACM International Conference On Computer Aided Design (ICCAD)},
     pages={1--9},
     year={2025},
     organization={IEEE}
   }