PQCache: Product Quantization-based KVCache for Long Context LLM Inference

Summary: PQCache uses PQ to compress KVCache for long-context LLM inference, treating KVCache as embedding retrieval. During prefilling and autoregressive decoding, PQ codes and centroids approximate key selection to fetch K/V, reducing overhead. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: h872d1978ad555b03
Venue: SIGMOD
Year: 2025
Pagerank: 6.8744954e-05
Overall Rank: 3,972 | 73.38%
DOI: 10.1145/3725338

Incoming Non-self Citations Over Time

Authors

1. Hailin Zhang (Peking University)
2. Xiaodong Ji (Peking University)
3. Yilin Chen (Peking University)
4. Fangcheng Fu (Peking University)
5. Xupeng Miao (Purdue University)
6. Xiaonan Nie (Peking University)
7. Weipeng Chen (Baichuan Inc.)
8. Bin Cui (Peking University)

BibTeX Citation

@inproceedings{zhang_sigmod25,
        title = {{PQCache: Product Quantization-based KVCache for Long Context LLM Inference}},
        author = {Zhang, Hailin and Ji, Xiaodong and Chen, Yilin and Fu, Fangcheng and Miao, Xupeng and Nie, Xiaonan and Chen, Weipeng and Cui, Bin},
        series = {{SIGMOD} '25},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3725338},
        url = {https://dl.acm.org/doi/10.1145/3725338},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 8 of 8 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
4,910	Apt-Serve: Adaptive Request Scheduling on Hybrid Cache for Scalable LLM Inference Serving	2025	SIGMOD	6.3468554e-05
5,376	RetroInfer: A Vector Storage Engine for Scalable Long-Context LLM Inference	2026	VLDB	6.139877e-05
6,455	AlayaDB: The Data Foundation for Efficient and Effective Long-context LLM Inference	2025	SIGMOD	5.762002e-05
6,944	DepCache: A KV Cache Management Framework for GraphRAG with Dependency Attention	2026	SIGMOD	5.6219388e-05
10,386	Buoy: Efficient and Effective Cache Replacement for Prefix Caching	2027	SIGMOD	4.9605343e-05
10,410	StaleFlow: Staleness-Aware Data Management for Mitigating Data Skewness in Fully Disaggregated RL Post-Training	2027	SIGMOD	4.9605343e-05
10,549	KVDrive: A Holistic Multi-Tier KV Cache Management System for Long-Context LLM Inference	2026	SIGMOD	4.9605343e-05
10,586	TaCo: Data-adaptive and Query-aware Subspace Collision for High-dimensional Approximate Nearest Neighbor Search	2026	SIGMOD	4.9605343e-05

Outgoing Citations (Sorted by Pagerank)

Showing 10 of 10 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
75	Fast Approximate Nearest Neighbor Search With The Navigating Spreading-out Graph	2019	VLDB	0.00037163157
710	Cache locality is not enough: High-Performance Nearest Neighbor Search with Product Quantization Fast Scan	2016	VLDB	0.00014598403
802	RaBitQ: Quantizing High-Dimensional Vectors with a Theoretical Error Bound for Approximate Nearest Neighbor Search	2024	SIGMOD	0.00013822222
814	SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data	2017	VLDB	0.00013697533
2,268	ELPIS: Graph-Based Similarity Search for Scalable Data Science	2023	VLDB	8.7144632e-05
2,426	DeltaPQ: Lossless Product Quantization Code Compression for High Dimensional Similarity Search	2020	VLDB	8.4692316e-05
7,066	Effective and Efficient Retrieval of Structured Entities	2020	VLDB	5.5928342e-05
7,590	Scalable Distributed Inverted List Indexes in Disaggregated Memory	2024	SIGMOD	5.4737949e-05
8,688	Accelerating Product Quantization Query Execution Runtime	2021	SIGMOD	5.2730113e-05
9,737	Experimental Analysis of Large-scale Learnable Vector Storage Compression	2024	VLDB	5.1187805e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	10,718	From Prefix Cache to Fusion RAG Cache: Accelerating LLM Inference in Retrieval-Augmented Generation	2026	SIGMOD
2	10,485	AlignedServe: Orchestrating Prefix-aware Batching to Build a High-throughput and Computing-efficient LLM Serving System	2026	SIGMOD
3	9,844	ContextCache: Context-Aware Semantic Cache for Multi-Turn Queries in Large Language Models	2025	VLDB
4	6,944	DepCache: A KV Cache Management Framework for GraphRAG with Dependency Attention	2026	SIGMOD
5	3,844	HotPrefix: Hotness-Aware KV Cache Scheduling for Efficient Prefix Sharing in LLM Inference Systems	2026	SIGMOD
6	5,376	RetroInfer: A Vector Storage Engine for Scalable Long-Context LLM Inference	2026	VLDB
7	10,389	CatKV: Accelerating Position-Independent Caching with Context-Adaptive KV Cache Compression	2027	SIGMOD
8	2,197	Cache-Craft: Managing Chunk-Caches for Efficient Retrieval-Augmented Generation	2025	SIGMOD
9	10,549	KVDrive: A Holistic Multi-Tier KV Cache Management System for Long-Context LLM Inference	2026	SIGMOD
10	10,847	Efficient Cooperation-Aware Key and Value Management for LLM Inference	2026	VLDB