DBScholar

Back to papers

Cache-Craft: Managing Chunk-Caches for Efficient Retrieval-Augmented Generation

Summary: Cache-Craft uses chunk-caches to reuse K/Vs in RAG, reducing repeated attention. It spots reusable chunk-caches, performs targeted recomputation to preserve quality, and hardware eviction to maximize reuse; reports 51% recomputation reduction and up to 2x throughput on LLaMA models. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID
7270
Venue
SIGMOD
Year
2025
Pagerank
8.0959781e-05
Overall Rank
2,816 | 80.69%
DOI
10.1145/3725273

Incoming Non-self Citations Over Time

Authors

BibTeX Citation

@inproceedings{agarwal_sigmod25,
        title = {{Cache-Craft: Managing Chunk-Caches for Efficient Retrieval-Augmented Generation}},
        author = {Agarwal, Shubham and Sundaresan, Sai and Mitra, Subrata and Mahapatra, Debabrata and Gupta, Archit and Sharma, Rounak and Kapu, Nirmal Joshua and Yu, Tong and Saini, Shiv},
        series = {{SIGMOD} '25},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3725273},
        url = {https://dl.acm.org/doi/10.1145/3725273},
        year = {2025}
}

Incoming Citations (Sorted by Pagerank)

Showing 6 of 6 citing papers.

Rank Citing Paper Year Venue Pagerank
6,254 In-depth Analysis of Graph-based RAG in a Unified Framework 2025 VLDB 5.9405174e-05
10,191 AlignedServe: Orchestrating Prefix-aware Batching to Build a High-throughput and Computing-efficient LLM Serving System 2026 SIGMOD 5.093636e-05
10,210 CoDec: Prefix-Shared Decoding Kernel for LLMs 2026 SIGMOD 5.093636e-05
10,290 Skyline Retrieval meets Set-Cover Chunk Merging: A Cost-Effective RAG-Sketch for Long-Context LLM QA 2026 SIGMOD 5.093636e-05
10,357 DepCache: A KV Cache Management Framework for GraphRAG with Dependency Attention 2026 SIGMOD 5.093636e-05
10,459 From Prefix Cache to Fusion RAG Cache: Accelerating LLM Inference in Retrieval-Augmented Generation 2026 SIGMOD 5.093636e-05
Previous Page 1 / 1 Next

Outgoing Citations (Sorted by Pagerank)

Showing 7 of 7 cited papers.

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

Rank Cited Paper Year Venue Pagerank
363 Approximate Query Processing: Taming the TeraBytes! A Tutorial 2001 VLDB 0.0002005475
772 VerdictDB: Universalizing Approximate Query Processing 2018 SIGMOD 0.00014147905
857 SlimDB: A Space-Efficient Key-Value Storage Engine For Semi-Sorted Data 2017 VLDB 0.00013568886
3,265 New Trends in High-D Vector Similarity Search: AI-driven, Progressive, and Distributed 2021 VLDB 7.5834255e-05
5,542 Put an Elephant into a Fridge: Optimizing Cache Efficiency for In-memory Key-value Stores 2020 VLDB 6.180117e-05
6,093 High-Dimensional Vector Similarity Search: From Time Series to Deep Network Embeddings 2020 SIGMOD 5.978635e-05
7,263 Catalyst: Optimizing Cache Management for Large In-memory Key-value Systems 2023 VLDB 5.6611756e-05
Previous Page 1 / 1 Next

Semantically Similar Papers

# Overall Rank Paper Year Venue
1 10,191 AlignedServe: Orchestrating Prefix-aware Batching to Build a High-throughput and Computing-efficient LLM Serving System 2026 SIGMOD
2 6,432 RetroInfer: A Vector Storage Engine for Scalable Long-Context LLM Inference 2026 VLDB
3 6,765 AquaPipe: A Quality-Aware Pipeline for Knowledge Retrieval and Large Language Models 2025 SIGMOD
4 6,788 HotPrefix: Hotness-Aware KV Cache Scheduling for Efficient Prefix Sharing in LLM Inference Systems 2026 SIGMOD
5 10,259 KVDrive: A Holistic Multi-Tier KV Cache Management System for Long-Context LLM Inference 2026 SIGMOD
6 13,340 ContextCache: Context-Aware Semantic Cache for Multi-Turn Queries in Large Language Models 2025 VLDB
7 10,290 Skyline Retrieval meets Set-Cover Chunk Merging: A Cost-Effective RAG-Sketch for Long-Context LLM QA 2026 SIGMOD
8 4,430 PQCache: Product Quantization-based KVCache for Long Context LLM Inference 2025 SIGMOD
9 10,357 DepCache: A KV Cache Management Framework for GraphRAG with Dependency Attention 2026 SIGMOD
10 10,459 From Prefix Cache to Fusion RAG Cache: Accelerating LLM Inference in Retrieval-Augmented Generation 2026 SIGMOD