AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

AdaCache: Adaptive Caching and Context Augmentation for Efficient LLM Serving

Zeng Zihao, Siyi Li, Xinyu Yan, Lei Xiao, Wei Yang Bryan Lim

ICLR 2026 · Large language models › Retrieval and knowledge · approach: RAG robustness and evaluation

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add adaptive context augmentation (ablation)Prefix Caching · RAG QA, 3 modelsTTFT: 1.65×Improved
add adaptive context augmentation (ablation)CacheBlend · RAG QA, 3 modelsTTFT: 1.22×Improved
add cache-aware selective recomputation (hierarchical cache) to ACA (ablation)Prefix Caching + ACA · RAG QA, 3 modelsTTFT: 1.76×Improved
add cache-aware selective recomputation (hierarchical cache) to ACA (ablation)CacheBlend + ACA · RAG QA, 3 modelsTTFT: 1.23×Improved
increase top-k retrieval from 2 to 8 (ablation)Prefix Caching · RAG TTFT vs full recomputationTTFT improvement: 1.76× -> 1.13×Hurt
increase top-k retrieval from 2 to 8 (ablation)AdaCache · RAG TTFT vs full recomputationTTFT improvement: 2.93× -> 4.67×Improved
skewed (head-heavy) context-length distribution (MMLU, TriviaQA) vs flatter (MMLU-Pro, SuperGPQA) (ablation)AdaCache vs CacheBlend · MMLU/TriviaQA vs MMLU-Pro/SuperGPQATTFT reduction: 1.95× and 1.62× vs 1.25× and 1.14×Improved
adaptive retrieval depth instead of static top-6ACA · top-6 retrieval costprefill tokens: 3,200 -> 1,280Improved
use AdaCache instead of full recomputationLLM RAG serving · 6 QA datasetsaccuracyNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy