Zicong Hong

发表论文 73 篇 · 总被引 2088 次 · h-index 19

代表论文

• Klotski: Efficient Mixture-of-Expert Inference via Expert-Aware Multi-Batch Pipeline (2025 · International Conference on Architectural Support for Programming Languages and Operating Systems · 被引 35)
• D2MoE: Dual Routing and Dynamic Scheduling for Efficient On-Device MoE-based LLM Serving (2025 · ACM/IEEE International Conference on Mobile Computing and Networking · 被引 26)
• Mell: Memory-Efficient Large Language Model Serving via Multi-GPU KV Cache Management (2025 · IEEE Conference on Computer Communications · 被引 13)
• The MoE-Empowered Edge LLMS Deployment: Architecture, Challenges, and Opportunities (2025 · IEEE Communications Magazine · 被引 11)
• Fate: Fasss sEsdge Inference of Mixture-of-Experts Models via Cross-Layer Gate (2025 · The Web Conference · 被引 10)
• HALO: A Unified Vision-Language-Action Model for Embodied Multimodal Chain-of-Thought Reasoning (2026 · arXiv.org · 被引 9)