6 papers
HBF Sucks! A Full-Stack Characterization of High-Bandwidth Flash for KV-Centric LLM Serving
Zhuoran Li, Zhuohang Bian, Xin Huang +3
A faster storage device should make serving faster. We find the opposite. High-Bandwidth Flash (HBF) stacks NAND behind a wide, package-local interface, promising flash-scale capac…
TokenCake: A KV-Cache-centric Serving Framework for LLM-based Multi-Agent Applications
Zhuohang Bian, Feiyang Wu, Zhuoran Li +2
Large Language Models (LLMs) are increasingly deployed in complex multi-agent applications that rely on external function calls. This workload creates severe performance challenges…
TokenStack: A Heterogeneous HBM-PIM Architecture and Runtime for Efficient LLM Inference
Zhuoran Li, Zhuohang Bian, Zihao Huang +5
Large language model (LLM) serving is now limited by the key-value (KV) cache. During decode, each new token rereads prior KV state, so attention becomes a bandwidth- and capacity-…
TokenDance: Scaling Multi-Agent LLM Serving via Collective KV Cache Sharing
Zhuohang Bian, Feiyang Wu, Chengrui Zhang +3
Multi-agent LLM applications organize execution in synchronized rounds where a central scheduler gathers outputs from all agents and redistributes the combined context. This All-Ga…
SLOTH: Lightweight Detection and Localization of On-Chip Fail-Slow Failures for DNN Accelerators
Junchi Wu, Xinfei Wan, Zhuoran Li +5
Spatial DNN accelerators are essential for high-performance inference, but their performance is undermined by widespread fail-slow failures. Detecting such failures on-chip is chal…
Mitigating Ambiguities in 3D Classification with Gaussian Splatting
Ruiqi Zhang, Hao Zhu, Jingyi Zhao +3
3D classification with point cloud input is a fundamental problem in 3D vision. However, due to the discrete nature and the insufficient material description of point cloud represe…