Wei Zhang

发表论文 13 篇 · 总被引 63 次 · h-index 4

代表论文

• PromptSculptor: Multi-Agent Based Text-to-Image Prompt Optimization (2025 · Conference on Empirical Methods in Natural Language Processing · 被引 34)
• Selective KV-Cache Sharing to Mitigate Timing Side-Channels in LLM Inference (2025 · arXiv.org · 被引 23)
• MCaM : Efficient LLM Inference with Multi-tier KV Cache Management (2025 · IEEE International Conference on Distributed Computing Systems · 被引 16)
• ExpertFlow: Adaptive Expert Scheduling and Memory Coordination for Efficient MoE Inference (2025 · arXiv.org · 被引 8)
• Dynamic Expert Quantization for Scalable Mixture-of-Experts Inference (2025 · arXiv.org · 被引 7)
• ACRFence: Preventing Semantic Rollback Attacks in Agent Checkpoint-Restore (2026 · arXiv.org · 被引 5)