VLCache: Computing 2% Vision Tokens and Reusing 98% for Vision-Language Inference

• URL: http://arxiv.org/abs/2512.12977v2
• Date: Thu, 18 Dec 2025 02:59:27 GMT
• Title: VLCache: Computing 2% Vision Tokens and Reusing 98% for Vision-Language Inference
• Authors: Shengling Qin, Hao Yu, Chenxin Wu, Zheng Li, Yizhong Cao, Zhengyang Zhuge, Yuxin Zhou, Wentao Yao, Yi Zhang, Zhengheng Wang, Shuai Bai, Jianwei Zhang, Junyang Lin,
• Abstract summary: VLCache is a cache reuse framework that exploits both KeyValue (KV) cache and encoderLang inputs to eliminate costly recomputation when the same multimodal cache recurs from prior approaches.<n>We show that VLCache achieves an accuracy on par with full recomputation, while requiring only 2-5% of the tokens to compute, yielding 1.2x-16x TTFT speedups.
• Score: 32.33685370786451
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: This paper presents VLCache, a cache reuse framework that exploits both Key-Value (KV) cache and encoder cache from prior multimodal inputs to eliminate costly recomputation when the same multimodal inputs recur. Unlike previous heuristic approaches, we formally identify the cumulative reuse error effect and demonstrate how to minimize the non-prefix cache reuse error effectively. We further analyze the varying importance of model layers and propose a dynamic, layer-aware recomputation strategy to balance accuracy and efficiency. Experimental results show that VLCache achieves an accuracy on par with full recomputation, while requiring only 2-5% of the tokens to compute, yielding 1.2x-16x TTFT speedups. We develop an experimental implementation of the proposed VLCache pipeline based on SGLang, enabling significantly faster inference in practical deployments.

Related papers

• SenCache: Accelerating Diffusion Model Inference via Sensitivity-Aware Caching [75.02865981328509]
  Caching reduces computation by reusing previously computed model outputs across timesteps.<n>We propose Sensitivity-Aware Caching (SenCache), a dynamic caching policy that adaptively selects caching timesteps on a per-sample basis.<n>SenCache achieves better visual quality than existing caching methods under similar computational budgets.
  arXiv  Detail & Related papers  (2026-02-27T17:36:09Z)
• ERTACache: Error Rectification and Timesteps Adjustment for Efficient Diffusion [30.897215456167753]
  Diffusion models suffer from substantial computational overhead due to their inherently iterative inference process.<n>We propose ERTACache, a principled caching framework that jointly rectifies both error types.<n>ERTACache achieves up to 2x inference speedup while consistently preserving or even improving visual quality.
  arXiv  Detail & Related papers  (2025-08-27T10:37:24Z)
• DiCache: Let Diffusion Model Determine Its Own Cache [62.954717254728166]
  DiCache is a training-free adaptive caching strategy for accelerating diffusion models at runtime.<n>Online Probe Profiling Scheme leverages a shallow-layer online probe to obtain an on-the-fly indicator for the caching error in real time.<n> Dynamic Cache Trajectory Alignment approximates the deep-layer feature output from multi-step historical caches.
  arXiv  Detail & Related papers  (2025-08-24T13:30:00Z)
• FastCache: Fast Caching for Diffusion Transformer Through Learnable Linear Approximation [43.83288560196838]
  Diffusion Transformers (DiT) are powerful generative models but remain computationally intensive due to their iterative structure and deep transformer stacks.<n>FastCache is a hidden-state-level caching and compression framework that accelerates DiT inference.<n> Empirical evaluations across multiple DiT variants demonstrate substantial reductions in latency and memory usage.
  arXiv  Detail & Related papers  (2025-05-26T05:58:49Z)
• dKV-Cache: The Cache for Diffusion Language Models [53.85291644298835]
  Diffusion Language Models (DLMs) have been seen as a promising competitor for autoregressive language models.<n>We propose a KV-cache-like mechanism, delayed KV-Cache, for the denoising process of DLMs.<n>Our approach is motivated by the observation that different tokens have distinct representation dynamics throughout the diffusion process.
  arXiv  Detail & Related papers  (2025-05-21T17:32:10Z)
• VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Caching [23.52474883720957]
  Vision-Language-Action (VLA) models have demonstrated strong multi-modal reasoning capabilities, enabling direct action generation from visual perception and language instructions.<n>This paper introduces VLA-Cache, a training-free inference acceleration method that reduces computational overhead by adaptively caching and reusing static visual tokens across frames.
  arXiv  Detail & Related papers  (2025-02-04T09:48:14Z)
• XKV: Personalized KV Cache Memory Reduction for Long-Context LLM Inference [9.65524177141491]
  Large Language Model (LLM) inference generates output tokens one-by-one, leading to many redundant computations.<n> KV-Cache framework makes a compromise between time and space complexities.<n>Existing studies reduce memory consumption by evicting some of cached data that have less important impact on inference accuracy.<n>We show that customizing the cache size for each layer in a personalized manner can yield a significant memory reduction.
  arXiv  Detail & Related papers  (2024-12-08T11:32:08Z)
• PrefixKV: Adaptive Prefix KV Cache is What Vision Instruction-Following Models Need for Efficient Generation [97.41972925670508]
  Large vision-language models (LVLMs) incur significant computational and memory overhead during inference.<n>We present PrefixKV, where "Prefix" means the top-ranked KV based on importance rather than position in the original sequence.<n>Our method achieves the state-of-the-art performance compared with others.
  arXiv  Detail & Related papers  (2024-12-04T15:48:59Z)
• VL-Cache: Sparsity and Modality-Aware KV Cache Compression for Vision-Language Model Inference Acceleration [7.463830743649754]
  Vision-Language Models (VLMs) have demonstrated impressive performance across a versatile set of tasks. Key-Value (KV) cache encodes long visual contexts, such as images or videos. Existing KV cache compression methods are effective for Large Language Models (LLMs) We propose a novel KV cache compression recipe tailored for accelerating VLM inference.
  arXiv  Detail & Related papers  (2024-10-29T20:04:34Z)
• ThinK: Thinner Key Cache by Query-Driven Pruning [63.13363917871414]
  Large Language Models (LLMs) have revolutionized the field of natural language processing, achieving unprecedented performance across a variety of applications.<n>This paper focuses on the long-context scenario, addressing the inefficiencies in KV cache memory consumption during inference.<n>We propose ThinK, a novel query-dependent KV cache pruning method designed to minimize attention weight loss while selectively pruning the least significant channels.
  arXiv  Detail & Related papers  (2024-07-30T17:59:08Z)
• Efficient Inference of Vision Instruction-Following Models with Elastic Cache [76.44955111634545]
  We introduce Elastic Cache, a novel strategy for efficient deployment of instruction-following large vision-language models. We propose an importance-driven cache merging strategy to prune redundancy caches. For instruction encoding, we utilize the frequency to evaluate the importance of caches. Results on a range of LVLMs demonstrate that Elastic Cache not only boosts efficiency but also notably outperforms existing pruning methods in language generation.
  arXiv  Detail & Related papers  (2024-07-25T15:29:05Z)
• DeepCache: Accelerating Diffusion Models for Free [65.02607075556742]
  DeepCache is a training-free paradigm that accelerates diffusion models from the perspective of model architecture. DeepCache capitalizes on the inherent temporal redundancy observed in the sequential denoising steps of diffusion models. Under the same throughput, DeepCache effectively achieves comparable or even marginally improved results with DDIM or PLMS.
  arXiv  Detail & Related papers  (2023-12-01T17:01:06Z)

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.