MEDA: Dynamic KV Cache Allocation for Efficient Multimodal Long-Context Inference

• URL: http://arxiv.org/abs/2502.17599v2
• Date: Thu, 13 Mar 2025 04:04:08 GMT
• Title: MEDA: Dynamic KV Cache Allocation for Efficient Multimodal Long-Context Inference
• Authors: Zhongwei Wan, Hui Shen, Xin Wang, Che Liu, Zheda Mai, Mi Zhang,
• Abstract summary: MEDA is a dynamic layer-wise KV cache allocation method for efficient multimodal long-context inference.<n> MEDA achieves up to 72% KV cache memory reduction and 2.82 times faster decoding speed.
• Score: 15.895020720304656
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Long-context Multimodal Large Language Models (MLLMs) that incorporate long text-image and text-video modalities, demand substantial resources as their multimodal Key-Value (KV) caches grow with increasing input lengths, challenging inference efficiency. Existing methods for KV cache compression, in both text-only and multimodal LLMs, have neglected attention density variations across layers, thus often adopting uniform or progressive reduction strategies for layer-wise cache allocation. In this work, we propose MEDA, a dynamic layer-wise KV cache allocation method for efficient multimodal long-context inference. As its core, MEDA utilizes cross-modal attention entropy to determine the KV cache size at each MLLMs layer. Given the dynamically allocated KV cache size at each layer, MEDA also employs a KV pair selection scheme to identify which KV pairs to select and a KV pair merging strategy that merges the selected and non-selected ones to preserve information from the entire context. MEDA achieves up to 72% KV cache memory reduction and 2.82 times faster decoding speed, while maintaining or enhancing performance on various multimodal tasks in long-context settings, including multi-images and long-video scenarios. Our code is released at https://github.com/AIoT-MLSys-Lab/MEDA.

Related papers

• DBudgetKV: Dynamic Budget in KV Cache Compression for Ensuring Optimal Performance [125.81664663201282]
  We introduce a new KV cache compression method dubbed DBudgetKV.<n>It features an attention-based metric to signal when the remaining KV cache is unlikely to match the full-cache performance, then halting the pruning process.<n>Our method is easy to integrate within LLM inference, not only optimizing memory space, but also showing reduced inference time compared to existing methods.
  arXiv  Detail & Related papers  (2025-02-24T06:33:39Z)
• DynamicKV: Task-Aware Adaptive KV Cache Compression for Long Context LLMs [31.62076958302603]
  Existing KV cache compression methods enforce a fixed pattern, neglecting task-specific characteristics and reducing the retention of essential information.<n>We propose DynamicKV, a method that dynamically optimize token retention by adjusting the number of tokens retained at each layer to adapt to the specific task.<n>Our method retains only 1.7% of the KV cache size while achieving 85% of the Full KV cache performance on LongBench.
  arXiv  Detail & Related papers  (2024-12-19T13:28:42Z)
• 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)
• PrefixKV: Adaptive Prefix KV Cache is What Vision Instruction-Following Models Need for Efficient Generation [65.36715026409873]
  Key-value (KV) cache, necessitated by the lengthy input and output sequences, notably contributes to the high inference cost. We present PrefixKV, which reframes the challenge of determining KV cache sizes for all layers into the task of searching for the optimal global prefix configuration. Our method achieves the state-of-the-art performance compared with others.
  arXiv  Detail & Related papers  (2024-12-04T15:48:59Z)
• 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)
• Inference-Friendly Models With MixAttention [7.103010772135246]
  MixAttention combines sliding window attention, where only a small subset of recent tokens is stored in the KV cache, with KV cache sharing across layers. Our experiments demonstrate that MixAttention significantly reduces memory usage and improves inference speed without sacrificing model performance in both short and long-context tasks.
  arXiv  Detail & Related papers  (2024-09-23T13:37:25Z)
• 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. This paper focuses on the long-context scenario, addressing the inefficiencies in KV cache memory consumption during inference. 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)
• Model Tells You Where to Merge: Adaptive KV Cache Merging for LLMs on Long-Context Tasks [21.815661269986425]
  We propose a novel KV cache merging approach, called KVMerger, to achieve adaptive KV cache compression for long-context tasks. Our approach is inspired by the intriguing observation that key states exhibit high similarity at the token level within a single sequence. We conduct extensive experiments to demonstrate the effectiveness of KVMerger for long-context tasks under constrained memory budgets.
  arXiv  Detail & Related papers  (2024-07-11T12:50:42Z)
• LOOK-M: Look-Once Optimization in KV Cache for Efficient Multimodal Long-Context Inference [32.20654044142376]
  LOOK-M is a pioneering, fine-tuning-free approach that efficiently reduces the multimodal KV cache size. It achieves up to 1.5x faster decoding and also maintains or even enhances performance across a variety of long context multimodal tasks.
  arXiv  Detail & Related papers  (2024-06-26T07:44:24Z)
• PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling [53.08975547824068]
  We investigate whether attention-based information flow inside large language models (LLMs) is aggregated through noticeable patterns for long context processing. Our observations reveal that LLMs aggregate information through Pyramidal Information Funneling where attention is scattering widely in lower layers. Motivated by these insights, we developed Pyramid KV, a novel and effective KV cache compression method.
  arXiv  Detail & Related papers  (2024-06-04T07:51:30Z)

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.