# HeteroCache: A Dynamic Retrieval Approach to Heterogeneous KV Cache Compression for Long-Context LLM Inference

arXiv: https://arxiv.org/abs/2601.13684 · HTML: https://mlsanity.com/paper/2601.13684/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in KV cache optimization, one of 12 directions in Efficient LLM inference and serving.
- Builds on 17 of its 27 references that are in the map.
- First posted on arXiv 20 Jan 2026.

## Placed in

- Efficient LLM inference and serving / [KV cache optimization](https://mlsanity.com/inference/kv-cache-optimization/)

## Builds on (most cited first, 17 of 17)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization](https://mlsanity.com/paper/2401.18079/) (2024)
- [Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference](https://mlsanity.com/paper/2406.10774/) (2024)
- [PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling](https://mlsanity.com/paper/2406.02069/) (2024)
- [8-bit Optimizers via Block-wise Quantization](https://mlsanity.com/paper/2110.02861/) (2021)
- [MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention](https://mlsanity.com/paper/2407.02490/) (2024)
- [LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks](https://mlsanity.com/paper/2412.15204/) (2024)
- [∞Bench: Extending Long Context Evaluation Beyond 100K Tokens](https://mlsanity.com/paper/2402.13718/) (2024)
- [FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving](https://mlsanity.com/paper/2501.01005/) (2025)
- [Data Engineering for Scaling Language Models to 128K Context](https://mlsanity.com/paper/2402.10171/) (2024)
- [ShadowKV: KV Cache in Shadows for High-Throughput Long-Context LLM Inference](https://mlsanity.com/paper/2410.21465/) (2024)
- [U-NIAH: Unified RAG and LLM Evaluation for Long Context Needle-In-A-Haystack](https://mlsanity.com/paper/2503.00353/) (2025)
- [LongBench Pro: A More Realistic and Comprehensive Bilingual Long-Context Evaluation Benchmark](https://mlsanity.com/paper/2601.02872/) (2026)

## Built on by (most cited first, 3 of 3)

- [Does Accuracy Equal Evidence? Reasoning Faithfulness under KV Cache Compression](https://mlsanity.com/paper/2608.01631/) (2026)
- [Adaptive Mass-Segmented KV Compression for Long-Context Reasoning](https://mlsanity.com/paper/2605.23200/) (2026)
- [WakeKV: Reactive, Reversible KV Residency for Heads That Change Their Minds](https://mlsanity.com/paper/2610.02713/) (2026)

## Abstract (arXiv)

The linear memory growth of the KV cache poses a significant bottleneck for LLM inference in long-context tasks. Existing static compression methods often fail to preserve globally important information. Although recent dynamic retrieval approaches attempt to address this issue, they typically suffer from coarse-grained caching strategies and incur high I/O overhead. To overcome these limitations, we propose HeteroCache, a training-free dynamic compression framework. Our method is built on two key insights: attention heads exhibit diverse temporal heterogeneity, and there is significant spatial redundancy among heads within the same layer. Guided by these insights, HeteroCache categorizes heads based on stability and similarity, applying a fine-grained weighting strategy that allocates larger cache budgets to heads with rapidly shifting attention to capture context changes. Furthermore, it features a hierarchical storage mechanism where representative heads monitor attention drift to trigger asynchronous, on-demand context retrieval, thereby hiding I/O latency. Experiments demonstrate that HeteroCache achieves state-of-the-art performance on long-context benchmarks and accelerates decoding by up to 3× compared to the original model with a 224K context. Our code is available at https://github.com/ponytaill/HeteroCache.