BitRed: Taming Non-Uniform Bit-Level Sparsity with a Programmable RISC-V ISA for DNN Acceleration.

Tianyu Liu 0007 刘天雨 Chinese Academy of Sciences, Institute of Computing Technology, Beijing, China University of Chinese Academy of Sciences, Beijing, China https://akaliu.github.io/ https://orcid.org/0000-0003-3905-6936 https://scholar.google.com/citations?user=xKcz2LYAAAAJ https://ieeexplore.ieee.org/author/37089385972 https://github.com/AkaLiu Tianyu Liu Tian-Yu Liu Tianyu Liu 0001 Alibaba Group, Qwen Team, Beijing, China Tencent, Cloud AI, Beijing, China Peking University, Key Laboratory of Computational Linguistics, Beijing, China https://tyliupku.github.io/ https://scholar.google.com/citations?user=6hHbBwwAAAAJ https://www.linkedin.com/in/tianyu-liu-28013a142/ https://twitter.com/rogerliuty Tianyu Liu 0004 Tianyu Liu 0002 University of Wisconsin-Madison, Madison, WI, USA https://scholar.google.com/citations?user=-FRI3MkAAAAJ Tianyu Liu 0003 University of Duisburg-Essen, Duisburg, Germany Harbin Institute of Technology, Department of Control Science and Engineering, Harbin, China https://www.linkedin.com/in/tianyu-liu-104492358/ https://orcid.org/0000-0002-8117-1201 https://scholar.google.com/citations?user=2lEcCQ8AAAAJ https://ieeexplore.ieee.org/author/37086570669 Tianyu Liu 0004 ETH Zürich, Switzerland Peking University, Beijing, China https://lyutyuh.github.io/ https://rycolab.io/authors/tianyu/ https://orcid.org/0000-0002-1147-8427 https://scholar.google.com/citations?user=tDQ2E0UAAAAJ https://www.linkedin.com/in/tianyu-liu-413391176/ https://twitter.com/t_y_liu https://github.com/lyutyuh Tianyu Liu 0001 Tianyu Liu 0005 Yale University, Department of Biostatistics, New Haven, CT, USA University of Illinois Urbana-Champaign, Department of Electrical and Computer Engineering, Urbana, IL, USA https://helloworldlty.github.io/cv/ https://orcid.org/0000-0002-9412-6573 https://scholar.google.com/citations?user=GdkHHjkAAAAJ https://openreview.net/profile?id=~Tianyu_Liu4 https://twitter.com/meSuper8 https://github.com/HelloWorldLTY Tianyu Liu 0006 Shandong Normal University, School of Information Science and Engineering, Jinan, China Shandong Management University, School of Business Administration, Jinan, China https://orcid.org/0000-0003-0774-8663 Tianyu Liu 0008 Hong Kong University of Science and Technology, Department of Electronic and Computer Engineering, Hong Kong Shenzhen Unity-Drive Innovation Technology Co. Ltd. (UDI), Shenzhen, China https://orcid.org/0000-0002-9128-5042 https://scholar.google.com/citations?user=NAt3vgcAAAAJ https://ieeexplore.ieee.org/author/37089508552 Tianyu Liu 0009 Shanghai Maritime University, School of Information Engineering, China https://orcid.org/0000-0003-0741-4743 Yanhuan Liu Wenming Li Kunming Zhang Yuqun Liu Siao Wen Lexin Wang Tianyu Liu 0007 Haibin Wu Zhihua Fan Xiaochun Ye Dongrui Fan Xuejun An BitRed: Taming Non-Uniform Bit-Level Sparsity with a Programmable RISC-V ISA for DNN Acceleration. 239-254 2026 ASPLOS (2) https://doi.org/10.1145/3779212.3790132 conf/asplos/2026-2 db/conf/asplos/asplos2026-2.html#LiuLZLWW0WFYFA26

Yanhuan Liu Wenming Li Kunming Zhang Tianyu Liu 0007 Xiaochun Ye Xuejun An CODA: A Computation-Driven Paradigm for Sparse DNN Acceleration. 381-384 2025 July - December 24 IEEE Comput. Archit. Lett. 2 https://doi.org/10.1109/LCA.2025.3637756 db/journals/cal/cal24.html#LiuLZLYA25 streams/journals/cal

Wenming Li Zhihua Fan Tianyu Liu 0007 Zhen Wang 0045 Haibin Wu Meng Wu 0006 Kunming Zhang Yanhuan Liu Ninghui Sun Xiaochun Ye Dongrui Fan DFU-E: A Dataflow Architecture for Edge DSP and AI Applications. 1100-1114 2025 June 36 IEEE Trans. Parallel Distributed Syst. 6 https://doi.org/10.1109/TPDS.2025.3555329 db/journals/tpds/tpds36.html#LiFLWWWZLSYF25 streams/journals/tpds

Yuhang Qiu Wenming Li Tianyu Liu 0007 Zhen Wang 0045 Zhiyuan Zhang Zhihua Fan Xiaochun Ye Dongrui Fan Zhimin Tang Accelerating Authenticated Block Ciphers via RISC-V Custom Cryptography Instructions. 1-7 2025 DATE https://doi.org/10.23919/DATE64628.2025.10992864 conf/date/2025 db/conf/date/date2025.html#QiuLLWZFYFT25

Zhihua Fan Wenming Li Zhen Wang 0045 Tianyu Liu 0007 Haibin Wu Yanhuan Liu Meng Wu 0006 Xinxin Wu Xiaochun Ye Dongrui Fan Ninghui Sun Xuejun An Accelerating Convolutional Neural Networks by Exploiting the Sparsity of Output Activation. 3253-3265 2023 December 34 IEEE Trans. Parallel Distributed Syst. 12 https://doi.org/10.1109/TPDS.2023.3324934 db/journals/tpds/tpds34.html#FanLWLWLWWYFSA23

Duo Wang Mingyu Yan Xin Liu 0073 Mo Zou Tianyu Liu 0007 Wenming Li Xiaochun Ye Dongrui Fan A High-accurate Multi-objective Exploration Framework for Design Space of CPU. 1-6 2023 DAC https://doi.org/10.1109/DAC56929.2023.10247790 conf/dac/2023 db/conf/dac/dac2023.html#WangYLZLLYF23 Shantian Qin Wenming Li Zhihua Fan Zhen Wang 0045 Tianyu Liu 0007 Haibin Wu Kunming Zhang Xuejun An Xiaochun Ye Dongrui Fan ROMA: A Reconfigurable On-chip Memory Architecture for Multi-core Accelerators. 49-57 2023 HPCC/DSS/SmartCity/DependSys https://doi.org/10.1109/HPCC-DSS-SmartCity-DependSys60770.2023.00017 conf/hpcc/2023 db/conf/hpcc/hpcc2023.html#QinLFWLWZAYF23 Tianyu Liu 0007 Wenming Li Zhihua Fan DFGC: DFG-aware NoC Control based on Time Stamp Prediction for Dataflow Architecture. 432-439 2023 ICCD https://doi.org/10.1109/ICCD58817.2023.00071 conf/iccd/2023 db/conf/iccd/iccd2023.html#LiuLF23 Haibin Wu Wenming Li Zhihua Fan Zhen Wang 0045 Tianyu Liu 0007 Junying Huang Shengzhong Tang Yanhuan Liu Kunming Zhang Xiaochun Ye Dongrui Fan Alleviating Transfer Latency in DataFlow Accelerator for DSP Applications. 440-443 2023 ICCD https://doi.org/10.1109/ICCD58817.2023.00073 conf/iccd/2023 db/conf/iccd/iccd2023.html#WuLFWLHTLZYF23

Yinshen Wang Wenming Li Tianyu Liu 0007 Liangjiang Zhou Bingnan Wang Zhihua Fan Xiaochun Ye Dongrui Fan Chibiao Ding Characterization and Implementation of Radar System Applications on a Reconfigurable Dataflow Architecture. 121-124 2022 July - December 21 IEEE Comput. Archit. Lett. 2 https://doi.org/10.1109/LCA.2022.3215595 db/journals/cal/cal21.html#WangLLZWFYFD22

Xinxin Wu Zhihua Fan Tianyu Liu 0007 Wenming Li Xiaochun Ye Dongrui Fan LRP: Predictive output activation based on SVD approach for CNN s acceleration. 831-836 2022 DATE https://doi.org/10.23919/DATE54114.2022.9774744 conf/date/2022 db/conf/date/date2022.html#WuFLLYF22 Zhihua Fan Wenming Li Tianyu Liu 0007 Shengzhong Tang Zhen Wang 0045 Xuejun An Xiaochun Ye Dongrui Fan A Loop Optimization Method for Dataflow Architecture. 202-211 2022 HPCC/DSS/SmartCity/DependSys https://doi.org/10.1109/HPCC-DSS-SmartCity-DependSys57074.2022.00059 conf/hpcc/2022 db/conf/hpcc/hpcc2022.html#FanLLTWAYF22 Zhihua Fan Wenming Li Tianyu Liu 0007 Xuejun An Xiaochun Ye Dongrui Fan A Routing-Aware Mapping Method for Dataflow Architectures. 3-16 2022 NPC https://doi.org/10.1007/978-3-031-21395-3_1 conf/npc/2022 db/conf/npc/npc2022.html#FanLLAYF22 Xuejun An Chibiao Ding Dongrui Fan Zhihua Fan Junying Huang Wenming Li Xin Liu 0073 Yanhuan Liu Yuqun Liu Shantian Qin Yuhang Qiu Ninghui Sun Shengzhong Tang Zhimin Tang Bingnan Wang Duo Wang Lexin Wang Yinshen Wang Zhen Wang 0045 Siao Wen Haibin Wu Meng Wu 0006 Xinxin Wu Mingyu Yan Xiaochun Ye Kunming Zhang Zhiyuan Zhang Liangjiang Zhou Mo Zou