📝 Selected Publications

You can also find my articles on my Google Scholar profile

{=} denotes equal contribution; {*} denotes corresponding author

TACO 2025 (Top Jour. in Computer Architecture)	Shiyuan Huang, Fangxin Liu=,, Zongwu Wang, Ning Yang, Haomin Li, Haibing Guan, and Li Jiang MIX-PC: Enabling Efficient DNN Inference with Mixed Numeric Precision Compilation Optimization (CCF Tier A)*
ASP-DAC 2026 (Top Conf. in Design Automation)	Zongwu Wang, Zhongyi Tang, Fangxin Liu, Chenyang Guan, Li Jiang, Haibing Guan TFLOP: Towards Energy-Efficient LLM Inference via An FPGA-Affinity Accelerator with Unified LUT-based Optimization (Acceptance Rate: 29%)
EMNLP 2025 (Top Conf. in NLP)	Fangxin Liu=, Zongwu Wang=, Jinhong Xia, Junping Zhao, Shouren Zhao, Jinjin Li, Jian Liu, Li Jiang, Haibing Guan FlexQuant: A Flexible and Efficient Dynamic Precision Switching Framework for LLM Quantization (Acceptance Rate: 22%) [Applied at Ant Group]
ICCAD 2025 (Top Conf. in Design Automation)	Yiwei Hu, Fangxin Liu=, Zongwu Wang, Yilong Zhao, Tao Yang, Haibing Guan, and Li Jiang PLAIN: Leveraging High Internal Bandwidth in PIM for Accelerating Large Language Model Inference via Mixed-Precision Quantization (Acceptance Rate: 24%)*
DAC 2025 (Top Conf. in Design Automation)	Zongwu Wang=, Peng Xu=, Fangxin Liu, Yiwei Hu, Qingxiao Sun, Gezi Li, Cheng Li, Xuan Wang, Li Jiang, and Haibing Guan MILLION: MasterIng Long-Context LLM InferenceVia Outlier-Immunized KV Product OuaNtization (Acceptance Rate: 23%)* [Applied at HUAWEI]
DAC 2025 (Top Conf. in Design Automation)	Ning Yang, Zongwu Wang, Qingxiao Sun, Liqiang Lu, and Fangxin Liu PISA:Efficient Precision-Slice Framework forLLMs with Adaptive Numerical Type (Acceptance Rate: 23%)*
DATE 2025 (Top Conf. in Design Automation)	Zongwu Wang, Fangxin Liu, Peng Xu, Qingxiao Sun, Junping Zhao and Li Jiang EVASION: Efficient KV Cache Compression via Product Quantization (Acceptance Rate: 21%)
ASP-DAC 2025 (Top Conf. in Design Automation)	Fangxin Liu=, Zongwu Wang=, Peng Xu, Shiyuan Huang and Li Jiang Exploiting Differential-Based Data Encoding for Enhanced Query Efficiency (Acceptance Rate: 28%)
ICCD 2024 (Import. Conf. in Computer Architecture)	Zongwu Wang=, Fangxin Liu=, and Li Jiang PS4:A Low Power SNN Accelerator with Spike Speculative Scheme (Acceptance Rate: 25%)
ICCD 2024 (Import. Conf. in Computer Architecture)	Longyu Zhao, Zongwu Wang, Fangxin Liu, and Li Jiang Ninja: A Hardware Assisted System for Accelerating Nested Address Translation (Acceptance Rate: 25%)*
MICRO 2024 (Top Conf. in Computer Architecture)	Zongwu Wang, Fangxin Liu, Ning Yang, Shiyuan Huang, Haomin Li, and Li Jiang COMPASS: SRAM-Based Computing-in-Memory SNN Accelerator with Adaptive Spike Speculation (Acceptance Rate: 22%)*
IEEE TPDS 2024 (Top Journal in Computer Architecture)	Fangxin Liu, Zongwu Wang, Wenbo Zhao, Ning Yang, Yongbiao Chen, Shiyuan Huang, Haomin Li, Tao Yang, Songwen Pei, Xiaoyao Liang,and Li Jiang Exploiting Temporal-Unrolled Parallelism for Energy-Efficient SNN Acceleration (CCF Tier A)
ISLPED 2024 (Top Conf. in Low Power Design)	Zongwu Wang, Fangxin Liu, Longyu Zhao, Shiyuan Huang and Li Jiang LowPASS: A Low power PIM-based accelerator with Speculative Scheme for SNNs (Acceptance Rate: 21%)*
ISCA 2024 (Top Conf. in Computer Architecture)	Yilong Zhao, Mingyu Gao, Fangxin Liu, Yiwei Hu, Zongwu Wang, Han Lin, Ji Li, He Xian, Hanlin Dong, Tao Yang, Naifeng Jing, Xiaoyao Liang, Li Jiang UM-PIM: DRAM-based PIM with Uniform & Shared Memory Space (Acceptance Rate: 18%)*
ICCD 2022	Fangxin Liu, Zongwu Wang, Yongbiao Chen, Li Jiang Randomize and Match: Exploiting Irregular Sparsity for Energy Efficient Processing in SNNs (Acceptance Rate: 24%)
IEEE TCAD 2022 (Top Journal in Computer-Aided Design)	Fangxin Liu, Zongwu Wang, Yongbiao Chen, Zhezhi He, Tao Yang, Xiaoyao Liang, Li Jiang SoBS-X: Squeeze-Out Bit Sparsity for ReRAM-Crossbar-Based Neural Network Accelerator (CCF Tier A)
DATE 2022 (Top Conf. in Design Automation)	Zongwu Wang, Zhezhi He, Rui Yang, Shiquan Fan, Jie Lin, Fangxin Liu, Yueyang Jia, Chenxi Yuan, Qidong Tang, Li Jiang. Self-Terminated Write of Multi-Level Cell ReRAM for Efficient Neuromorphic Computing (Best Paper Award)