# ZSMerge: Zero-Shot KV Cache Compression for Memory-Efficient Long-Context LLMs

arXiv: https://arxiv.org/abs/2503.10714 · HTML: https://mlsanity.com/paper/2503.10714/

## Key facts

- Cited by 8 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 41 of its 50 references that are in the map.
- First posted on arXiv 13 Mar 2025.

## Placed in

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

## Builds on (most cited first, 30 of 41)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [Linformer: Self-Attention with Linear Complexity](https://mlsanity.com/paper/2006.04768/) (2020)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints](https://mlsanity.com/paper/2305.13245/) (2023)
- [Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization](https://mlsanity.com/paper/1808.08745/) (2018)
- [Token Merging: Your ViT But Faster](https://mlsanity.com/paper/2210.09461/) (2022)
- [SnapKV: LLM Knows What You are Looking for Before Generation](https://mlsanity.com/paper/2404.14469/) (2024)
- [Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned](https://mlsanity.com/paper/1905.09418/) (2019)
- [H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models](https://mlsanity.com/paper/2306.14048/) (2023)
- [Yi: Open Foundation Models by 01.AI](https://mlsanity.com/paper/2403.04652/) (2024)
- [Fast Transformer Decoding: One Write-Head is All You Need](https://mlsanity.com/paper/1911.02150/) (2019)
- [The Falcon Series of Open Language Models](https://mlsanity.com/paper/2311.16867/) (2023)
- [Efficient Content-Based Sparse Attention with Routing Transformers](https://mlsanity.com/paper/2003.05997/) (2020)
- [PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling](https://mlsanity.com/paper/2406.02069/) (2024)
- [Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention](https://mlsanity.com/paper/2502.11089/) (2025)
- [SliceGPT: Compress Large Language Models by Deleting Rows and Columns](https://mlsanity.com/paper/2401.15024/) (2024)
- [MInference 1.0: Accelerating Pre-filling for Long-Context LLMs via Dynamic Sparse Attention](https://mlsanity.com/paper/2407.02490/) (2024)
- [∞Bench: Extending Long Context Evaluation Beyond 100K Tokens](https://mlsanity.com/paper/2402.13718/) (2024)
- [LooGLE: Can Long-Context Language Models Understand Long Contexts?](https://mlsanity.com/paper/2311.04939/) (2023)
- [MoBA: Mixture of Block Attention for Long-Context LLMs](https://mlsanity.com/paper/2502.13189/) (2025)
- [InfLLM: Training-Free Long-Context Extrapolation for LLMs with an Efficient Context Memory](https://mlsanity.com/paper/2402.04617/) (2024)
- [A Survey on Recent Advances in LLM-Based Multi-turn Dialogue Systems](https://mlsanity.com/paper/2402.18013/) (2024)

## Built on by (most cited first, 8 of 8)

- [Cartridges at Scale: Training Modular KV Caches over Large Document Collections](https://mlsanity.com/paper/2606.04557/) (2026)
- [Every Cache Entry Earns Its Place: Global Allocation of Resolution and Coverage for KV Cache Compression](https://mlsanity.com/paper/2608.07001/) (2026)
- [TPLA: Tensor Parallel Latent Attention for Efficient Disaggregated Prefill and Decode Inference](https://mlsanity.com/paper/2508.15881/) (2025)
- [Joint Encoding of KV-Cache Blocks for Scalable LLM Serving](https://mlsanity.com/paper/2601.03067/) (2026)
- [Key-Value Means: Transformers with Expandable Block-Recurrent Compressed Memory](https://mlsanity.com/paper/2605.09877/) (2026)
- [KVSlimmer: Theoretical Insights and Practical Optimizations for Asymmetric KV Merging](https://mlsanity.com/paper/2603.00907/) (2026)
- [Meta-Soft: Leveraging Composable Meta-Tokens for Context-Preserving KV Cache Compression](https://mlsanity.com/paper/2605.22337/) (2026)
- [VisCache: Visual KV Cache Pruning for Efficient Vision Large Language Model Inference](https://mlsanity.com/paper/2608.24063/) (2026)

## Abstract (arXiv)

The linear growth of key-value (KV) cache memory and quadratic computational in attention mechanisms complexity pose significant bottlenecks for large language models (LLMs) in long-context processing. While existing KV cache optimization methods address these challenges through token pruning or feature merging, they often incur irreversible information loss or require costly parameter retraining. To this end, we propose ZSMerge, a dynamic KV cache compression framework designed for efficient cache management, featuring three key operations: (1) fine-grained memory allocation guided by multi-dimensional token importance metrics at head-level granularity, (2) a residual merging mechanism that preserves critical context through compensated attention scoring, and (3) a zero-shot adaptation mechanism compatible with diverse LLM architectures without requiring retraining. ZSMerge significantly enhances memory efficiency and inference speed with negligible performance degradation across LLMs. When applied to LLaMA2-7B, it demonstrates a 20:1 compression ratio for key-value cache retention (reducing memory footprint to 5% of baseline) while sustaining comparable generation quality, coupled with triple throughput gains at extreme 54k-token contexts that eliminate out-of-memory failures. The code is available at https://github.com/SusCom-Lab/ZSMerge.