I am Wenyong Zhou (周文涌), a PhD graduate from The University of Hong Kong (HKU), where I had the privilege of being guided by Prof. Ngai Wong and Prof. Can Li in the Next Gen AI Lab. Before that, I earned a Bachelor’s degree from the School of Microelectronics at Tianjin University (2019), where I was fortunate to be mentored by Prof. Yugong Wu. I then pursued a Master’s degree in Electrical and Computer Engineering at Northwestern University, where I was deeply influenced by the visionary mentorship of Prof. Seda Ogrenci.

My research focuses on implicit neural representations (INRs) as an efficient data paradigm and emerging computational frameworks, including compute-in-memory (CIM) and stochastic computing. Additionally, I am passionate about improving the efficiency of Large Language Models (LLMs) through techniques such as quantization, pruning, and knowledge distillation.

Since November 2024, I have been working at Zhicun (Witmem) Technology, focusing on low-bit training of LLMs, particularly for compatibility with analog CIM hardware. Previously, I enjoyed enriching internship experiences at Bytedance and JD.com in 2021 and 2023, respectively.

Research Interests

• Data is the fuel of the AI era, and INRs offer an innovative approach for encoding data such as images, signals, and 3D scenes in a continuous and compact form using neural networks. This reduces the need for extensive storage while maintaining high-fidelity representations.

• Computing power drives modern AI, but classical digital computers face challenges due to the separation of compute and storage units. CIM integrates memory and computation into a single unit, significantly reducing data movement and making it especially effective for accelerating AI workloads.

• LLMs have become a cornerstone of modern AI, driving advancements in applications ranging from natural language understanding to content generation. Minimizing their size and computational requirements without compromising performance democratizes access to advanced AI capabilities and broadens their range of applications.

Recent News

• 2025.07 - One paper was accepted by IEEE TCAD.
• 2025.06 - One paper was accepted by ICCV 2025.
• 2025.06 - One paper was accepted by IEEE TCAS-II.
• 2025.03 - One paper was accepted by ICME 2025.
• 2024.12 - Two papers were accepted by ICCASP 2025.
• 2024.11 - One paper was accepted by DATE 2025.

Selected Publications

*: Equal contribution.

• INRs
  • W. Zhou*, B. Li*, T. Wu, C. Ding, Z. Liu and N. Wong. QuadINR: Quadratic Implicit Neural Representations for Efficient Memristor-based CIM System, IEEE Transactions on Circuits and Systems II: Express Briefs.
  • J. Ren*, W. Zhou*, T. Wu, Y. Cheng, Z. Liu and N. Wong. Patch-Based Implicit Neural Representations for Efficient and Scalable Inference, IEEE Transactions on Circuits and Systems II: Express Briefs, (Major revision).
  • W. Zhou*, J. Ren*, T. Wu, Y. Cheng, Z. Liu and N. Wong. Distribution-Aware Hadamard Quantization for Hardware-Efficient Implicit Neural Representations, 2025 IEEE International Conference on Multimedia and Expo (ICME), Nantes, France, 2025.
  • W. Zhou*, T. Wu*, Y. Cheng, C. Zhang, Z. Liu and N. Wong. MINR: Efficient Implicit Neural Representations for Multi-Image Encoding, ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Hyderabad, India, 2025.
  • W. Zhou*, Y. Cheng*, T. Wu, C. Zhang, Z. Liu and N. Wong. Enhancing Robustness of Implicit Neural Representations Against Weight Perturbations, *ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Hyderabad, India, 2025.
• CIM hardware
  • T. Wu*, C. Ding*, W. Zhou*, Y. Cheng, X. Feng, C. Shi, Z. Liu and N. Wong. HaLoRA: Hardware-aware Low-Rank Adaptation for Large Language Models Based on Hybrid Compute-in-Memory Architecture, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, (Under review).
  • W. Zhou, T. Wu, C. Ding, Y. Ren, Z. Liu and N. Wong. Towards RRAM-based Transformer-based Vision Models with Noise-aware Knowledge Distillation, 2025 Design, Automation & Test in Europe Conference & Exhibition (DATE), Lyon, France, 2025.
  • Y. Feng*, W. Zhou*, Y. Lyu, H. Liu, Z. Liu, N. Wong and W. Kang, HPD: Hybrid Projection Decomposition for Robust State Space Models on Analog CIM Hardware, 2025 IEEE 16th International Conference on ASIC (ASICON), Kunming, China, 2025.
  • W. Zhou, Y. Ren, Z. Liu and N. Wong. Binary Weight Multi-Bit Activation Quantization for Compute-in-Memory CNN Accelerators, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.
• Efficient LLMs
  • On the way

More about me

• I am passionate about competitive sports and enjoy practicing table tennis, badminton, and basketball. These activities have taught me valuable lessons about handling failure and shaping my personality.

• I enjoy reading, particularly exploring history and politics through a financial lens to uncover how fundamental financial principles remain unchanged despite human intentions.

• I love traveling the world to experience beautiful natural landscapes and diverse human cultures, which enrich my understanding of humanity and inspire new ways of thinking.

(Last updated on Aug., 2025)