The risk of KV cache compression

• URL: http://arxiv.org/abs/2607.01520v1
• Date: Wed, 01 Jul 2026 22:36:21 GMT
• Title: The risk of KV cache compression
• Abstract summary: Transformer inference on long sequences is expensive because softmax attention repeatedly reads from a large KV cache.<n>The prevalent approach to this bottleneck is KV cache compression, which replaces the full cache with a compact summary.<n>We bridge this gap by characterizing the minimax risk of KV cache compression in terms of the intrinsic compressibility of a cache, revealing when and how accurate compression is possible.
• Score: 27.1081027038941
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Transformer inference on long sequences is expensive because softmax attention repeatedly reads from a large KV cache. The prevalent approach to this bottleneck is KV cache compression, which replaces the full cache with a compact summary. Despite its practical importance, the design of such summaries is largely driven by empirical experimentation. On the theoretical side, existing results show that KV cache compression can be impossible in the worst case, but offer little systematic guidance for designing algorithms in regimes where accurate compression is possible. We bridge this gap by characterizing the minimax risk of KV cache compression in terms of the intrinsic compressibility of a cache, revealing when and how accurate compression is possible. These results yield novel design principles for KV cache compression under causal masking that map efficiently to prefill and autoregressive decoding while achieving minimax-optimal risk. We instantiate these principles in a practical algorithm and report promising performance on LongBench in targeted experiments. Overall, our results provide a principled avenue for practical KV cache compression with theoretical guarantees.

Related papers

• EchoKV: Efficient KV Cache Compression via Similarity-Based Reconstruction [55.026048429595384]
  EchoKV is a flexible KV cache compression scheme that enables on-demand transitions between standard and compressed inference.<n>We introduce a two-stage fine-tuning strategy that allows for rapid, low-cost training.
  arXiv  Detail & Related papers  (2026-03-24T07:58:42Z)
• One Size Does Not Fit All: Token-Wise Adaptive Compression for KV Cache [38.49582847975703]
  We propose DynaKV, a novel post-training framework for low-rank KV cache compression.<n>Our method consistently outperforms existing state-of-the-art compression techniques.<n>When integrated with SnapKV, DynaKV retains only 6% of the KV cache while maintaining 94% of the baseline performance on the LongBench benchmark.
  arXiv  Detail & Related papers  (2026-02-03T13:20:36Z)
• More Than a Quick Glance: Overcoming the Greedy Bias in KV-Cache Compression [0.0]
  LASER-KV is a framework designed to test the limits of KV compression under a strict accumulative budgeting policy.<n>Experiments on the Babilong benchmark reveal performance degradation in previous compression methods by 15-30% on various long context tasks.
  arXiv  Detail & Related papers  (2026-02-02T15:05:03Z)
• The Pitfalls of KV Cache Compression [52.196873305708955]
  We show that certain instructions degrade much more rapidly with compression.<n>We show several factors that play a role in prompt leakage: compression method, instruction order, and KV eviction bias.
  arXiv  Detail & Related papers  (2025-09-30T19:55:26Z)
• KVComp: A High-Performance, LLM-Aware, Lossy Compression Framework for KV Cache [7.019967158501771]
  We present KVComp, a generic and efficient KV cache management framework optimized for long-text generation.<n> KVComp employs novel lossy compression techniques specifically designed for KV cache data characteristics.<n>We show that KVComp achieves on average 47% and up to 83% higher memory reduction rate compared to existing methods.
  arXiv  Detail & Related papers  (2025-08-30T18:25:19Z)
• CommonKV: Compressing KV Cache with Cross-layer Parameter Sharing [54.34080239841088]
  CommonKV is a training-free method for cross-layer KV cache compression through adjacent parameters sharing.<n>We show that the proposed method consistently outperforms existing low-rank and cross-layer approaches at various compression ratios.
  arXiv  Detail & Related papers  (2025-08-22T06:55:45Z)
• ReCalKV: Low-Rank KV Cache Compression via Head Reordering and Offline Calibration [69.57122277845293]
  We propose ReCalKV, a post-training low-rank KV cache compression approach with tailored strategies for Keys and Values.<n>For Keys, we propose Similarity aware Recontext (HSR), which clusters structurally similar heads into groups, enabling more accurate low-rank approximation.<n>For Values, we propose Offline Head-wise Value (OVC), which efficiently calibrates the value projection matrix using calibration data without training.
  arXiv  Detail & Related papers  (2025-05-30T08:49:27Z)
• DBudgetKV: Dynamic Budget in KV Cache Compression for Ensuring Optimal Performance [125.81664663201282]
  We introduce a new KV cache compression method dubbed DBudgetKV.<n>It features an attention-based metric to signal when the remaining KV cache is unlikely to match the full-cache performance.<n>Our method achieves lossless KV pruning effectively and robustly, exceeding 25% compression ratio on average.
  arXiv  Detail & Related papers  (2025-02-24T06:33:39Z)
• KVSharer: Efficient Inference via Layer-Wise Dissimilar KV Cache Sharing [58.29726147780976]
  We propose a plug-and-play method called textit KVSharer, which shares the KV cache between layers to achieve layer-wise compression. Experiments show that textit KVSharer can reduce KV cache computation by 30%, thereby lowering memory consumption. We verify that textit KVSharer is compatible with existing intra-layer KV cache compression methods, and combining both can further save memory.
  arXiv  Detail & Related papers  (2024-10-24T08:06:41Z)
• LoRC: Low-Rank Compression for LLMs KV Cache with a Progressive Compression Strategy [59.1298692559785]
  Key-Value ( KV) cache is crucial component in serving transformer-based autoregressive large language models (LLMs) Existing approaches to mitigate this issue include: (1) efficient attention variants integrated in upcycling stages; (2) KV cache compression at test time; and (3) KV cache compression at test time. We propose a low-rank approximation of KV weight matrices, allowing plug-in integration with existing transformer-based LLMs without model retraining. Our method is designed to function without model tuning in upcycling stages or task-specific profiling in test stages.
  arXiv  Detail & Related papers  (2024-10-04T03:10:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.