Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

RaBitQCache: Rotated Binary Quantization for KVCache in Long Context LLM Inference

About

Long-context Large Language Model inference is severely bottlenecked by the massive Key-Value (KV) cache, yet existing sparse attention methods often suffer from static fixed-budget (Top-k) retrieval or rely on proxy scores that are computationally expensive and biased. To address these limitations, we propose RaBitQCache, a novel sparse attention framework that utilizes randomized rotated binary quantization and high-throughput binary-INT4 arithmetic to efficiently estimate attention weights. Our proxy score serves as an unbiased estimator with a proven error bound, enabling adaptive Top-p retrieval that dynamically adjusts the token budget based on actual attention sparsity. We further implement a hardware-aware system with asynchronous pipelining and lazy updates to mask overhead. Evaluations demonstrate that RaBitQCache significantly accelerates inference and reduces memory I/O while preserving generation quality compared to state-of-the-art baselines. Code is available at https://github.com/Sakuraaa0/RaBitQCache.git.

Wenhao Li, Jinhao Dong, Hailin Zhang, Wenhang Shi, Wei Lu, Xiaoyong Du• 2026

Related benchmarks

TaskDatasetResultRank
Long-context Language UnderstandingLongBench
Average Score54.58
90
Long-context language modelingRULER
Accuracy (8K Context)84.7
80
Long-context language modelingLongBench
Generation Score50.6
24
Mathematical ReasoningGSM8K
Accuracy77
13
Long-context Language UnderstandingLongBench
Average Score50.63
6
Mathematical ReasoningGSM8K
Accuracy (GSM8K)77
4
Synthetic Long-context RetrievalRULER 8K–64K
Average Accuracy79.55
4
Showing 7 of 7 rows

Other info

Follow for update