Pith. sign in

REVIEW 19 cited by

GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2403.05527 v4 pith:S3677HHV submitted 2024-03-08 cs.LG cs.AIcs.CL

classification cs.LGcs.AIcs.CL
keywords gearcachecompressionentriesinferencenear-losslessachievesefficient
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Key-value (KV) caching has become the de-facto to accelerate generation speed for large language models (LLMs) inference. However, the growing cache demand with increasing sequence length has transformed LLM inference to be a memory bound problem, significantly constraining the system throughput. Existing methods rely on dropping unimportant tokens or quantizing all entries uniformly. Such methods, however, often incur high approximation errors to represent the compressed matrices. The autoregressive decoding process further compounds the error of each step, resulting in critical deviation in model generation and deterioration of performance. To tackle this challenge, we propose GEAR, an efficient KV cache compression framework that achieves near-lossless high-ratio compression. GEAR first applies quantization to majority of entries of similar magnitudes to ultra-low precision. It then employs a low rank matrix to approximate the quantization error, and a sparse matrix to remedy individual errors from outlier entries. By adeptly integrating three techniques, GEAR is able to fully exploit their synergistic potentials. Our experiments demonstrate that compared to alternatives, GEAR achieves near-lossless 4-bit KV cache compression with up to 2.38x throughput improvement, while reducing peak-memory size up to 2.29x. Our code is publicly available at https://github.com/HaoKang-Timmy/GEAR.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 19 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Runtime Observability for Heterogeneous Attention Memory

    cs.AI 2026-08 conditional novelty 7.0 of 10

    A runtime observability framework with typed, Lean-checked error contracts for four kinds of attention memory, tested on six LLM serving setups and a production corruption case.

  2. PolarQuant: Quantizing KV Caches with Polar Transformation

    cs.LG 2025-02 conditional novelty 7.0 of 10

    PolarQuant achieves around 4x KV cache compression for LLMs by quantizing angles after a recursive polar transform with random preconditioning, with LongBench scores close to the full-precision model and above prior c...

  3. From Tensor Buffer to Distributed Memory Hierarchy: A Survey of KV Cache Management for LLM Serving

    cs.DC 2026-06 accept novelty 6.5 of 10

    KV-cache serving systems concentrate into five archetypes under a four-axis taxonomy, with ownership explaining residual distributed design variance and seven measurement gaps blocking next steps.

  4. Practical Online KV Cache Compaction for LLM Agents: An Empirical Study

    cs.CL 2026-08 conditional novelty 6.0 of 10

    In LLM agent trajectories, delaying KV cache compaction by a few turns and using the agent's later queries as the compression guide preserves most accuracy while cutting KV cache by 80%.

  5. What to Keep, What to Forget: A Rate--Distortion View of Memory Compaction in LLMs and Agents

    cs.LG 2026-07 conditional novelty 6.0 of 10

    KV-cache eviction, prompt compression, recurrent state bounding, and agent memory consolidation are unified as one rate-distortion problem with a shared lower bound, shared failure mode, and transferable mechanisms.

  6. CompressKV: Semantic Retrieval Heads Know What Tokens are Not Important Before Generation

    cs.CL 2025-08 unverdicted novelty 6.0 of 10

    A method that uses a handful of 'semantic retrieval heads' instead of all attention heads to decide which key-value cache entries can be dropped, plus layer-wise cache budgeting, reportedly beats prior KV compression ...

  7. CaliDrop: KV Cache Compression with Calibration

    cs.CL 2025-07 conditional novelty 6.0 of 10

    CaliDrop adds a stale-query calibration term on top of token eviction, improving accuracy at high KV compression ratios with modest throughput overhead.

  8. TailorKV: A Hybrid Framework for Long-Context Inference via Tailored KV Cache Optimization

    cs.CL 2025-05 conditional novelty 6.0 of 10

    TailorKV combines 1-bit quantization in shallow attention layers with dynamic Top-K token retrieval in deeper layers to serve 128k-context Llama-3.1-8B on a single 24GB GPU with a small accuracy loss.

  9. Inference-time sparse attention with asymmetric indexing

    cs.CL 2025-02 conditional novelty 6.0 of 10

    Saap approximates attention by clustering keys with k-means and learning a query classifier, reducing KV-cache lookups about 20x at 4-5% selectivity with small accuracy loss on several long-context benchmarks.

  10. SHARP: Accelerating Language Model Inference by SHaring Adjacent layers with Recovery Parameters

    cs.LG 2025-02 conditional novelty 6.0 of 10

    Sharing one MLP layer's weights across several layers plus low-rank adapters recovers most of a pretrained LLM's quality with a fraction of the storage and faster phone inference.

  11. QuantSpec: Self-Speculative Decoding with Hierarchical Quantized KV Cache

    cs.LG 2025-02 conditional novelty 6.0 of 10

    QuantSpec accelerates long-context LLM decoding by giving the self-draft model a 4-bit hierarchical KV cache shared with an INT8-quality target cache, achieving up to ~2.5x speedup and mostly >90% acceptance.

  12. PolarQuant: Leveraging Polar Transformation for Efficient Key Cache Quantization and Decoding Acceleration

    cs.LG 2025-02 conditional novelty 6.0 of 10

    Quantizing LLM key caches in polar coordinates after RoPE preserves accuracy at lower total bit width and turns query-key multiplication into a lookup-table operation.

  13. Cache Me If You Must: Adaptive Key-Value Quantization for Large Language Models

    cs.LG 2025-01 conditional novelty 6.0 of 10

    Trained linear predictors across transformer layers plus residual quantization compress LLM KV caches to about 2 bits per value with near-lossless accuracy on tested Llama and Qwen models.

  14. TaDA: Training-free recipe for Decoding with Adaptive KV Cache Compression and Mean-centering

    cs.CL 2025-06 conditional novelty 5.0 of 10

    TaDA mean-centers key and value activations along the head dimension, quantizes the deviations with per-layer precision search, and matches 16-bit baseline accuracy while using about 27 to 35 percent of the cache memory.

  15. Memory-Efficient Visual Autoregressive Modeling with Scale-Aware KV Cache Compression

    cs.LG 2025-05 conditional novelty 5.0 of 10

    ScaleKV cuts KV cache memory for Visual Autoregressive text-to-image generation to 10% by classifying layers as drafters or refiners per scale and pruning low-attention tokens while keeping benchmark scores nearly unchanged.

  16. HACK: Homomorphic Acceleration via Compression of the Key-Value Cache for Disaggregated LLM Inference

    cs.DC 2025-02 conditional novelty 5.0 of 10

    HACK performs attention matrix multiplications directly on 2-bit quantized key-value cache entries, eliminating per-iteration dequantization and cutting end-to-end latency in disaggregated LLM serving.

  17. Towards Efficient Large Language Model Serving: A Survey on System-Aware KV Cache Optimization

    cs.LG 2026-07 accept novelty 4.0 of 10

    A survey organizing serving-time KV cache optimization techniques into temporal, spatial, and structural system behaviors, analyzing cross-behavior co-design patterns and open challenges.

  18. PagedEviction: Structured Block-wise KV Cache Pruning for Efficient Large Language Model Inference

    cs.LG 2025-09 conditional novelty 4.0 of 10

    PagedEviction prunes the KV cache in whole blocks using a key-value norm ratio, speeding up long-context LLM inference in vLLM while staying close to full-cache accuracy.

  19. Win Fast or Lose Slow: Balancing Speed and Accuracy in Latency-Sensitive Decisions of LLMs

    cs.LG 2025-05 reject novelty 4.0 of 10

    Adaptively quantizing parts of an LLM's layers to FP4 can improve win rates and trading yields in latency-sensitive agent tasks, but the reported gains come from choosing the best compression level after seeing test results.

Pith tools