Meta-Soft: Leveraging Composable Meta-Tokens for Context-Preserving KV Cache Compression

• URL: http://arxiv.org/abs/2605.22337v2
• Date: Sat, 23 May 2026 16:54:41 GMT
• Title: Meta-Soft: Leveraging Composable Meta-Tokens for Context-Preserving KV Cache Compression
• Abstract summary: KV cache used in large language models has linearly growing time complexity.<n>Current KV Cache eviction methods rely on a static parameter set as the query to evaluate the importance of KV pairs.<n>We propose Meta-Soft, a dynamic compression framework based on probe-driven context integration.
• Score: 12.589754555104768
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The KV cache used in large language models has linearly growing time complexity, so LLMs face memory blow-up and reduced decoding efficiency when they process long contexts. Current KV Cache eviction has become an important research direction; however, existing methods based on fixed Soft Tokens (e.g., Judge Q) rely on a static parameter set as the query to evaluate the importance of KV pairs, so they cannot adapt dynamically to different input prompts, and they cannot precisely capture complex and changing task relevance. Also, evicted KV pairs are discarded permanently, so this causes irreversible information loss and context breaks. To address this problem, we propose Meta-Soft, a dynamic compression framework based on probe-driven context integration. Specifically, we build a meta-library with a learnable orthogonal basis matrix $\mathcal{L}$, and we use a selector network with Gumbel-Softmax to produce differentiable sparse combination weights, so we dynamically synthesize the most targeted $k$ Soft Tokens from the input prompt features. We append these Soft Tokens to the end of the input sequence to probe key information. We also introduce an attention-flow based integration mechanism, which redistributes the semantic information of removed tokens into retained tokens, and this keeps the dropped context information effectively. Experiments on multiple datasets show that our method outperforms existing state-of-the-art eviction methods and provides a new solution for KV Cache compression.

Related papers

• Back from the Future: Key-Value Cache Management by Counter-Causal Surprise [67.1056509495879]
  Key-value (KV) cache management through compression and eviction strategies has emerged as an important research direction in recent years.<n>We propose a simple yet effective KV eviction scheme motivated by the insight that past tokens which can be well-predicted from more recent tokens are redundant.<n>We evaluate our strategy on various open-source LLMs and benchmark datasets showing competitive or improved performance over other state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-07-30T02:42:51Z)
• SelKV: Selective KV Cache Merging with Per-Token Merge-or-Drop and Attention Compensation [0.0]
  Large Language Models generate text autoregressively, relying on a key-value (KV) cache whose memory footprint grows linearly with context length.<n>Recent compression methods mitigate this cost via token merging.<n>We propose a training-free, dual-component framework for KV cache compression that addresses these limitations.
  arXiv  Detail & Related papers  (2026-05-19T17:44:56Z)
• Detecting Overflow in Compressed Token Representations for Retrieval-Augmented Generation [49.48204107529758]
  We define token overflow as a regime in which compressed representations no longer contain sufficient information to answer a given query.<n>In this paper, we find that query-agnostic saturation statistics reliably separate compressed from uncompressed token representations.<n>Lightweight probing classifiers over both query and context xRAG representations detect overflow with 0.72 AUC-ROC on average.<n>These results advance from query-independent diagnostics to query-aware detectors, enabling low-cost pre-LLM gating to mitigate compression-induced errors.
  arXiv  Detail & Related papers  (2026-02-12T18:15:08Z)
• SemShareKV: Efficient KVCache Sharing for Semantically Similar Prompts via Token-Level LSH Matching [0.8307668828380427]
  We propose textitSemShareKV, a KV cache sharing and compression framework for large language models (LLMs)<n>Instead of relying on exact token matches, SemShareKV applies fuzzy token matching using locality-sensitive hashing (LSH) on token embeddings and incorporates Rotary Position Embedding (RoPE) to better preserve positional information.<n> Experiments on diverse summarization datasets show up to 6.25$times$ speedup and 42% lower GPU memory usage with 5k tokens input, with negligible quality degradation.
  arXiv  Detail & Related papers  (2025-09-29T14:16:13Z)
• UniGist: Towards General and Hardware-aligned Sequence-level Long Context Compression [86.33995240043936]
  UniGist is a sequence-level long-context compression framework for large language models.<n>It efficiently preserves context information by replacing raw tokens with special compression tokens (gists) in a fine-grained manner.<n>Our scheme also supports flexible inference by allowing the actual removal of compressed tokens, resulting in real-time memory savings.
  arXiv  Detail & Related papers  (2025-09-19T08:47:37Z)
• Judge Q: Trainable Queries for Optimized Information Retention in KV Cache Eviction [53.83828564664595]
  Large language models (LLMs) utilize key-value ( KV) cache to store historical information during sequence processing.<n>Current methods for KV cache eviction typically utilize the last window from the pre-filling phase as queries to compute the KV importance scores for eviction.<n>We propose Judge Q, a novel training method which incorporates a soft token list.
  arXiv  Detail & Related papers  (2025-09-13T03:34:12Z)
• DAST: Context-Aware Compression in LLMs via Dynamic Allocation of Soft Tokens [20.044306399439265]
  Large Language Models (LLMs) face computational inefficiencies and redundant processing when handling long context inputs.<n>We propose Dynamic Allocation of Soft Tokens (DAST), a simple yet effective method that leverages the LLM's intrinsic understanding of contextual relevance to guide compression.<n> Experimental results across multiple benchmarks demonstrate that DAST surpasses state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-02-17T06:55:13Z)
• 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)
• Compressing KV Cache for Long-Context LLM Inference with Inter-Layer Attention Similarity [24.118503938098307]
  textscPoD allocates memory according to token importance.<n>textscPoD reduces KV cache memory usage by up to 35% without compromising performance.
  arXiv  Detail & Related papers  (2024-12-03T08:29:27Z)
• 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)
• Context Compression for Auto-regressive Transformers with Sentinel Tokens [37.07722536907739]
  We propose a plug-and-play approach that is able to incrementally compress the intermediate activation of a specified span of tokens into compact ones. Experiments on both in-domain language modeling and zero-shot open-ended document generation demonstrate the advantage of our approach.
  arXiv  Detail & Related papers  (2023-10-12T09:18:19Z)

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.