SeKV: Resolution-Adaptive KV Cache with Hierarchical Semantic Memory for Long-Context LLM Inference

• URL: http://arxiv.org/abs/2606.31145v1
• Date: Tue, 30 Jun 2026 05:18:02 GMT
• Title: SeKV: Resolution-Adaptive KV Cache with Hierarchical Semantic Memory for Long-Context LLM Inference
• Abstract summary: SeKV is a resolution-adaptive semantic KV cache that organizes context into entropy-guided semantic spans.<n>We show that SeKV improves over the strongest semantic compression baseline by 5.9% on average.<n>It reduces GPU memory by 53.3% versus full KV caching at 128K context.
• Score: 37.37437232987781
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models increasingly operate over long contexts, where the KV cache becomes a dominant memory bottleneck: its size grows linearly with sequence length and must be retained throughout decoding, making full GPU caching prohibitively expensive without compression. Existing KV cache compression methods struggle to balance efficiency with faithful context preservation. Token eviction discards information, while semantic grouping fixes compression decisions at prefill time; neither can recover token-level detail from a compressed span once it becomes relevant during generation. As a solution, we propose SeKV, a resolution-adaptive semantic KV cache that organizes context into entropy-guided semantic spans and stores them across a GPU-CPU memory hierarchy without discarding information. Each span keeps a lightweight summary vector on GPU for coarse routing and a low-rank SVD basis on CPU for on-demand token-level reconstruction. A trained zoom-in mechanism selectively expands query-relevant spans during decoding, enabling precise retrieval without materializing the full KV cache on GPU. SeKV enables adaptive token-level reconstruction while keeping the base LLM fully frozen and adding fewer than 0.05% trainable parameters. Across four benchmarks, SeKV improves over the strongest semantic compression baseline by 5.9% on average while reducing GPU memory by 53.3% versus full KV caching at 128K context. Code is available on https://github.com/AmirAbaskohi/SeKV.

Related papers

• FlashMemory-DeepSeek-V4: Lightning Index Ultra-Long Context via Lookahead Sparse Attention [77.12062766962815]
  Lookahead Sparse Attention (LSA) is powered by a Neural Memory Indexer built upon the DeepSeek-V4 architecture.<n>We instantiate this architecture via a backbone-free decoupled training strategy.<n>We show that FM-DS-V4 compresses the average physical KV cache footprint down to merely 13.5% of the full-context baseline.
  arXiv  Detail & Related papers  (2026-06-08T06:25:54Z)
• 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)
• PackKV: Reducing KV Cache Memory Footprint through LLM-Aware Lossy Compression [8.427136461713706]
  We present textbfPackKV, a generic and efficient KV cache management framework.<n>PackKV supports both latency-critical and throughput-critical inference scenarios.
  arXiv  Detail & Related papers  (2025-12-30T20:05:32Z)
• KVReviver: Reversible KV Cache Compression with Sketch-Based Token Reconstruction [20.53279247581787]
  We propose KVReviver, a reversible KV cache compression method based on the sketch algorithm.<n>In 2k-length contexts, it requires only 10% of KV Cache budget while maintaining identical end-to-end inference accuracy.<n>For 32k-length contexts, it achieves equivalent or comparable accuracy 2% accuracy loss.
  arXiv  Detail & Related papers  (2025-12-01T03:59:20Z)
• StreamKV: Streaming Video Question-Answering with Segment-based KV Cache Retrieval and Compression [95.59657871147846]
  We propose textbfStreamKV, a framework that seamlessly equips Video-LLMs with advanced KV cache retrieval and compression.<n>Experiments on public StreamingVQA benchmarks demonstrate that StreamKV significantly outperforms existing Online Video-LLMs.
  arXiv  Detail & Related papers  (2025-11-10T16:25:03Z)
• KV Cache Transform Coding for Compact Storage in LLM Inference [2.20003167536462]
  We present KVTC, a lightweight transform coder that compresses KV caches for compact on- GPU and off- GPU storage.<n>By exploiting redundancies in KV caches, KVTC achieves up to 20$times$ compression while maintaining reasoning and long-context accuracy.<n>We test KVTC with Llama 3, Mistral NeMo, and R1-Qwen 2.5 models across benchmarks including AIME25, LiveCodeBench, GSM8K, MMLU, Qasper, RULER, and MATH-500.
  arXiv  Detail & Related papers  (2025-11-03T18:20:35Z)
• SpeCache: Speculative Key-Value Caching for Efficient Generation of LLMs [44.41154292836592]
  We propose SpeCache, which offloads the complete KV cache and dynamically fetches KV pairs back in each decoding step.<n> Experiments on LongBench and Needle-in-a-Haystack benchmarks verify that SpeCache effectively reduces VRAM usage.
  arXiv  Detail & Related papers  (2025-03-20T14:01:56Z)
• ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference [61.412894960600205]
  Large Language Models (LLMs) require significant GPU memory when processing long texts.<n>ChunkKV reimagines KV cache compression by treating semantic chunks as basic compression units.<n>Result: ChunkKV outperforms state-of-the-art methods by up to 8.7% in precision.
  arXiv  Detail & Related papers  (2025-02-01T03:49:47Z)
• SCBench: A KV Cache-Centric Analysis of Long-Context Methods [61.025422435235456]
  We introduce SCBench, a benchmark for evaluating long-context methods from a KV cachecentric perspective.<n>We provide an extensive KV cache-centric analysis of eight categories long-context solutions, including Gated Linear RNNs and Mamba-Attention hybrids.<n>Our findings show that sub-O(n) memory methods suffer in multi-turn scenarios, while sparse encoding with O(n) memory and sub-O(n2) pre-filling perform robustly.
  arXiv  Detail & Related papers  (2024-12-13T17:59:52Z)
• 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.