Yue Zhu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yue Zhu has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Chen Wang
AI Cloud Platform IBM T.J. Watson Research Center Yorktown Heights NY USA
3 shared publications - Huamin Chen
IBM Research Red Hat Yorktown Heights NY Boston MA
2 shared publications Xunzhuo Liu
IBM Research Tencent Yorktown Heights NY
2 shared publications- Alaa Youssef
IBM Research IBM Research Yorktown Heights NY Yorktown Heights NY
1 shared publication - Eun Kyung Lee
IBM T. J. Watson Research Center Yorktown Heights NY USA
1 shared publication - Hao Yu
IBM T. J. Watson Research Center Yorktown Heights NY USA
1 shared publication
Latest Publications
- Open Access
Category-Aware Semantic Caching for Heterogeneous LLM Workloads
LLM serving systems process heterogeneous query workloads where different categories exhibit different characteristics. Code queries cluster densely in embedding space while conversational queries distribute sparsely. Content staleness varies from minutes (stock data) to months (code patterns). Query repetition patterns range from power-law (code) to uniform (conversation), producing long tail cache hit rate distributions: high-repetition categories achieve 40-60% hit rates while low-repetition or volatile categories achieve 5-15% hit rates. Vector databases must exclude the long tail because remote search costs (30ms) require 15--20% hit rates to break even, leaving 20-30% of production traffic uncached. Uniform cache policies compound this problem: fixed thresholds cause false positives in dense spaces and miss valid paraphrases in sparse spaces; fixed TTLs waste memory or serve stale data. This paper presents category-aware semantic caching where similarity thresholds, TTLs, and quotas vary by query category. We present a hybrid architecture separating in-memory HNSW search from external document storage, reducing miss cost from 30ms to 2ms. This reduction makes low-hit-rate categories economically viable (break-even at 3-5% versus 15-20%), enabling cache coverage across the entire workload distribution. Adaptive load-based policies extend this framework to respond to downstream model load, dynamically adjusting thresholds and TTLs to reduce traffic to overloaded models by 9-17% in theoretical projections.
Open AccessCategory-Aware Semantic Caching for Heterogeneous LLM Workloads
LLM serving systems process heterogeneous query workloads where different categories exhibit different characteristics. Code queries cluster densely in embedding space while conversational queries distribute sparsely. Content staleness varies from minutes (stock data) to months (code patterns). Query repetition patterns range from power-law (code) to uniform (conversation), producing long tail cache hit rate distributions: high-repetition categories achieve 40-60% hit rates while low-repetition or volatile categories achieve 5-15% hit rates. Vector databases must exclude the long tail because remote search costs (30ms) require 15--20% hit rates to break even, leaving 20-30% of production traffic uncached. Uniform cache policies compound this problem: fixed thresholds cause false positives in dense spaces and miss valid paraphrases in sparse spaces; fixed TTLs waste memory or serve stale data. This paper presents category-aware semantic caching where similarity thresholds, TTLs, and quotas vary by query category. We present a hybrid architecture separating in-memory HNSW search from external document storage, reducing miss cost from 30ms to 2ms. This reduction makes low-hit-rate categories economically viable (break-even at 3-5% versus 15-20%), enabling cache coverage across the entire workload distribution. Adaptive load-based policies extend this framework to respond to downstream model load, dynamically adjusting thresholds and TTLs to reduce traffic to overloaded models by 9-17% in theoretical projections.
- Open Access
When to Reason: Semantic Router for vLLM
Large Language Models (LLMs) demonstrate substantial accuracy gains when augmented with reasoning modes such as chain-of-thought and inference-time scaling. However, reasoning also incurs significant costs in inference latency and token usage, with environmental and financial impacts, which are unnecessary for many simple prompts. We present a semantic router that classifies queries based on their reasoning requirements and selectively applies reasoning only when beneficial. Our approach achieves a 10.2 percentage point improvement in accuracy on the MMLU-Pro benchmark while reducing response latency by 47.1% and token consumption by 48.5% compared to direct inference with vLLM. These results demonstrate that semantic routing offers an effective mechanism for striking a balance between accuracy and efficiency in open-source LLM serving systems
Open AccessWhen to Reason: Semantic Router for vLLM
Large Language Models (LLMs) demonstrate substantial accuracy gains when augmented with reasoning modes such as chain-of-thought and inference-time scaling. However, reasoning also incurs significant costs in inference latency and token usage, with environmental and financial impacts, which are unnecessary for many simple prompts. We present a semantic router that classifies queries based on their reasoning requirements and selectively applies reasoning only when beneficial. Our approach achieves a 10.2 percentage point improvement in accuracy on the MMLU-Pro benchmark while reducing response latency by 47.1% and token consumption by 48.5% compared to direct inference with vLLM. These results demonstrate that semantic routing offers an effective mechanism for striking a balance between accuracy and efficiency in open-source LLM serving systems
- Open Access
Towards Efficient Key-Value Cache Management for Prefix Prefilling in LLM Inference
The increasing adoption of large language models (LLMs) with extended context windows necessitates efficient Key-Value Cache (KVC) management to optimize inference performance. Inference workloads like Retrieval-Augmented Generation (RAG) and agents exhibit high cache reusability, making efficient caching critical to reducing redundancy and improving speed. We analyze real-world KVC access patterns using publicly available traces and evaluate commercial key-value stores like Redis and state-of-the-art RDMA-based systems (CHIME [1] and Sherman [2]) for KVC metadata management. Our work demonstrates the lack of tailored storage solution for KVC prefilling, underscores the need for an efficient distributed caching system with optimized metadata management for LLM workloads, and provides insights into designing improved KVC management systems for scalable, low-latency inference.
Open AccessTowards Efficient Key-Value Cache Management for Prefix Prefilling in LLM Inference
The increasing adoption of large language models (LLMs) with extended context windows necessitates efficient Key-Value Cache (KVC) management to optimize inference performance. Inference workloads like Retrieval-Augmented Generation (RAG) and agents exhibit high cache reusability, making efficient caching critical to reducing redundancy and improving speed. We analyze real-world KVC access patterns using publicly available traces and evaluate commercial key-value stores like Redis and state-of-the-art RDMA-based systems (CHIME [1] and Sherman [2]) for KVC metadata management. Our work demonstrates the lack of tailored storage solution for KVC prefilling, underscores the need for an efficient distributed caching system with optimized metadata management for LLM workloads, and provides insights into designing improved KVC management systems for scalable, low-latency inference.