Minima-KV: Retention-Preserving KV Cache Compression with Mixed-Format Paged Attention

• URL: http://arxiv.org/abs/2608.23834v1
• Date: Mon, 24 Aug 2026 21:21:58 GMT
• Title: Minima-KV: Retention-Preserving KV Cache Compression with Mixed-Format Paged Attention
• Abstract summary: Minima-KV is a retention-preserving hierarchy for mixed-format paged attention.<n>Recent and protected Anchor pages remain in FP8, while older non-anchor pages move to packed TQ3.<n> format-specific kernels compute partial attention states and combine them through a globally normalized online-softmax state.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The key-value (KV) cache is a primary capacity and bandwidth bottleneck in long-context LLM serving. We present Minima-KV, a retention-preserving hierarchy for mixed-format paged attention. Recent and protected Anchor pages remain in FP8, while older non-anchor pages move to packed TQ3; every live-request page remains addressable. Format-specific kernels compute partial attention states and combine them through a globally normalized online-softmax merge, enabling direct heterogeneous decode without a cache-sized dense shadow. Across separate, configuration-bound Qwen3.6-27B profiles on a single 96-GB NVIDIA RTX PRO 6000 Blackwell GPU, deployment accounting reports 18.3 KiB of attention KV per live token, corresponding to 3.50x compression relative to BF16 and 1.75x relative to FP8. A materializing quality profile matches its dense control on 16K RULER needle-in-a-haystack tasks. On the same 503-question LongBench v2 set, measured deltas are -0.80, -0.60, and -0.40 percentage points at 16K, 32K, and 64K. A separate single-pair direct-decode canary with two 59,008-token requests measures 3.625x active-KV compression and 0.9821x throughput relative to its control, routes all 16 full-attention layers without fallback, and retains no dense shadow. These results establish a practical mixed-format path for compressing long-context state without evicting live-request KV pages.

Related papers

• PuzzleKV: Page-Wise Low-Rank Decomposition for KV Cache Compression [6.397490960170785]
  PuzzleKV is a low-rank compression method for large language models.<n>It decomposes pages within each layer and KV head, computes attention directly over dense and factorized pages, and incrementally compresses newly eligible pages.<n>It achieves more than 96% of Full KV performance across both evaluated models and all benchmark settings.
  arXiv  Detail & Related papers  (2026-08-24T21:30:11Z)
• FreqDepthKV: Frequency-Guided Depth Sharing for Robust KV Cache Compression in Long-Context LLM Inference [0.0]
  FreqDepthKV is an inference-time cache compression method that factorizes adjacent-layer KV states into shared low-frequency depth components and sparse high-frequency residuals.<n>A lightweight online probe assigns attention heads to shared-depth, residual-depth, or exact cache modes according to their contribution to reconstruction-sensitive attention logits.<n>Across long-context question answering, needle retrieval, summarization, and code generation benchmarks, FreqDepthKV preserves task accuracy under substantially smaller cache budgets.
  arXiv  Detail & Related papers  (2026-07-07T17:26:28Z)
• RoPE-Aware Bit Allocation for KV-Cache Quantization [52.099459337231345]
  Block-GTQ is a bit allocator for key-cache quantization built on TurboQuant-MSE.<n>It better preserves RoPE query-key logits on a ten-model diagnostic panel.<n>It runs 1.34x faster than fp16 FlashAttention2 at 128K context.
  arXiv  Detail & Related papers  (2026-06-23T00:17:48Z)
• KVSculpt: KV Cache Compression as Distillation [7.085426079187912]
  KV cache compression is critical for efficient long-context LLM inference.<n>Existing methods range from pure eviction -- selecting which KV pairs to keep -- to merging, which combines similar pairs into fewer ones.<n>We propose KVSculpt, which moves to the other end of this spectrum.<n>Keys are optimized via L-BFGS and values are solved in closed form via least squares.
  arXiv  Detail & Related papers  (2026-03-29T19:14:25Z)
• DeltaKV: Residual-Based KV Cache Compression via Long-Range Similarity [50.52392445266824]
  We propose a residual-based KV cache compression framework motivated by long-range inter-token similarity and highly shared latent components in KV representations.<n>Instead of discarding tokens, DeltaKV encodes semantic residuals relative to retrieved historical references, preserving fidelity while substantially reducing storage.<n>Experiments show that DeltaKV reduces KV cache memory to 29% of the original while maintaining near-lossless accuracy on LongBench, SCBench, and AIME.
  arXiv  Detail & Related papers  (2026-02-08T15:14:36Z)
• OjaKV: Context-Aware Online Low-Rank KV Cache Compression with Oja's Rule [54.37983890753086]
  We introduce OjaKV, a framework that integrates a strategic hybrid storage policy with online subspace adaptation.<n>OjaKV preserves crucial first and most recent tokens in full-rank, maintaining high-fidelity anchors for attention.<n>It applies low-rank compression by incrementally adapting the projection basis using Oja's algorithm for online principal component analysis.
  arXiv  Detail & Related papers  (2025-09-25T21:42:27Z)
• xKV: Cross-Layer SVD for KV-Cache Compression [8.250015628919098]
  Large Language Models (LLMs) with long context windows enable powerful applications but come at the cost of high memory consumption.<n>Recent studies attempted to merge KV-cache from multiple layers into shared representations.<n>We find that the dominant singular vectors are remarkably well-aligned across multiple layers of the KV-Cache.<n>xKV consolidates the KV-Cache of multiple layers into a shared low-rank subspace, significantly reducing KV-Cache sizes.
  arXiv  Detail & Related papers  (2025-03-24T17:06:37Z)
• 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)
• KV-Compress: Paged KV-Cache Compression with Variable Compression Rates per Attention Head [0.8158530638728501]
  We introduce KV-Compress, a novel compression method that evicts contiguous KV blocks within a PagedAttention framework. Our method achieves state-of-the-art performance on LongBench for both Mistral-7B-Instruct-v0.2 and Llama-3.1-8B-Instruct while lowering the total number of compressed KVs by 4x. Evaluations on Llama-3.1-8B-Instruct and Llama-3.1-70B-Instruct-FP8 achieve compression rates up to 8x with negligible impact on performance, and up to 64x while retaining over 90% of full-cache performance
  arXiv  Detail & Related papers  (2024-09-30T19:09:13Z)
• LoCoCo: Dropping In Convolutions for Long Context Compression [77.26610232994508]
  This paper presents a novel approach, Dropping In Convolutions for Long Context Compression (LoCoCo) LoCoCo employs only a fixed-size Key-Value ( KV) cache, and can enhance efficiency in both inference and fine-tuning stages.
  arXiv  Detail & Related papers  (2024-06-08T01:35:11Z)
• PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling [38.732413451399]
  Pyramid KV is a novel and effective KV cache compression method.<n>We show that Pyramid KV matches the performance of models with a full KV cache while retaining only 12% of the KV cache.<n>In the Needle-in-a-Haystack experiment, Pyramid KV outperforms competing methods in maintaining long-context comprehension.
  arXiv  Detail & Related papers  (2024-06-04T07:51:30Z)
• Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs [82.08922896531618]
  We introduce adaptive KV cache compression, a plug-and-play method that reduces the memory footprint of generative inference for Large Language Models (LLMs) We conduct targeted profiling to discern the intrinsic structure of attention modules. Based on the recognized structure, we then construct the KV cache in an adaptive manner: evicting long-range contexts on attention heads emphasizing local contexts, discarding non-special tokens on attention heads centered on special tokens, and only employing the standard KV cache for attention heads that broadly attend to all tokens.
  arXiv  Detail & Related papers  (2023-10-03T05:17:08Z)

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.