Year: 2026
Type: conference-paper
Abstract: Multimodal Large Language Models face severe challenges in computational efficiency and memory consumption due to the substantial expansion of the visual KV cache when processing long visual contexts.Existing KV cache compression methods typically rely on the "persistence of importance" hypothesis to prune tokens.However, this approach proves fragile in multimodal settings due to two key issues: 1) Visual tokens display "deferred importance," initially exhibiting low salience but becoming pivotal during later decoding, which can lead to premature eviction.2) Discrete pruning disrupts the inherent spatial continuity of visual cues.To address these challenges, we propose RetentiveKV, an entropydriven KV cache optimization method that reformulates KV eviction from "discrete context truncation" to "continuous memory evolution" based on State Space Models.Our method leverages information entropy to quantify the information potential of low-attention tokens and integrates tokens scheduled for eviction into a continuous state space through entropyguided state transitions, enabling their dynamic reactivation when semantic relevance arises during subsequent decoding.Extensive experiments on multimodal benchmarks demonstrate that RetentiveKV achieves 5.0 × KV cache compression and 1.5 × decoding acceleration.
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Awards: G1101031052, G2567805129, G5672621438
OpenAlex ID: W7166859219