一作MICRO 2026Oral· 3D-Stacked & PIM Architectures for LLM/MoE Serving
Gossamer: A Utility-Driven Architecture for Constant-Budget KV Cache Compression in Reasoning LLMs
Feng Cheng · Cong Guo · Junyao Zhang · Haoxuan Shan · Chiyue Wei · Hong Wang · Hai Helen Li · Yiran Chen