CVLGMar 15

ASAP: Attention-Shift-Aware Pruning for Efficient LVLM Inference

arXiv:2603.1454959.81 citationsh-index: 1
AI Analysis

This addresses efficiency issues for users of LVLMs, though it is incremental as it builds on existing token reduction strategies.

The paper tackled the computational bottleneck of processing high-resolution visual tokens in Large Vision-Language Models by proposing ASAP, a training-free pruning method that reduces FLOPs by ~80% while retaining 99.02% of the original model performance.

While Large Vision-Language Models (LVLMs) demonstrate exceptional multi-modal capabilities, the quadratic computational cost of processing high-resolution visual tokens remains a critical bottleneck. Though recent token reduction strategies attempt to accelerate inference, such methods inadequately exploit attention values and fail to address token redundancy. More critically, they overlook the ``attention shift'' phenomenon inherent in LVLMs, which skews token attention scores. In this work, we propose ASAP, a novel training-free, KV-Cache-compatible pruning recipe that comprehensively addresses these limitations. First, we mitigate the attention shift by utilizing a dynamic bidirectional soft attention mask, ensuring the selection of genuinely informative tokens rather than naive attention-based selection. Second, we posit that high semantic redundancy within the token set degrades performance. We therefore introduce a weighted soft merging component that merges semantically similar tokens, preserving only the most feature-dense visual patches for subsequent layers. ASAP achieves virtually lossless compression of visual context, retaining 99.02% of the original LLaVA-NeXT-7B performance while aggressively slashing computational FLOPs by ~80%.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes