5 papers · 1 filter
EfficientAgent: What Makes KV Cache Offloading Work for Concurrent Agents?
Kunming Shao, Jierun Chen, Jiangnan Yu +7
LLM agents resend their whole conversation on every turn, and most of it was already processed on the previous turn. Serving systems avoid recomputing it by caching its key-value (…
PQ-HSA: Reusing Product-Quantized Scores for Hybrid Sparse-Approximate Attention
Kunming Shao, Jierun Chen, Yanli Wang +4
At each decoding step a language model attends over the key-value (KV) cache of every earlier token, so at long context the attention call is bounded by memory bandwidth. Sparse at…
MoE Expert Execution in Disaggregated LLM Serving with a High-Bandwidth ReRAM Near-Memory Architecture
Kunming Shao, Ming Zeng, Xin Yuan +5
Attention-FFN disaggregation maps LLM modules to specialized pools, creating an opening to keep Mixture-of-Experts (MoE) weights resident in a high-bandwidth FFN pool. Decode SLOs,…
Balancing FP8 Computation Accuracy and Efficiency on Digital CIM via Shift-Aware On-the-fly Aligned-Mantissa Bitwidth Prediction
Liang Zhao, Kunming Shao, Zhipeng Liao +4
FP8 low-precision formats have gained significant adoption in Transformer inference and training. However, existing digital compute-in-memory (DCIM) architectures face challenges i…
DS-CIM: Digital Stochastic Computing-In-Memory Featuring Accurate OR-Accumulation via Sample Region Remapping for Edge AI Models
Kunming Shao, Liang Zhao, Jiangnan Yu +5
Stochastic computing (SC) offers hardware simplicity but suffers from low throughput, while high-throughput Digital Computing-in-Memory (DCIM) is bottlenecked by costly adder logic…