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

• URL: http://arxiv.org/abs/2603.14549v2
• Date: Tue, 17 Mar 2026 20:41:37 GMT
• Title: ASAP: Attention-Shift-Aware Pruning for Efficient LVLM Inference
• Abstract summary: We propose ASAP, a novel training-free, KV-Cache-compatible pruning recipe.<n>We mitigate the attention shift by utilizing a dynamic bidirectional soft attention mask.<n>We also posit that high semantic redundancy within the token set degrades performance.
• Score: 26.71949723961501
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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%.

Related papers

• Capturing Token Tendencies for Training-Free Token Pruning in Multimodal Large Language Models [86.02826269126308]
  We propose Trend-aware Pruning, a novel framework that elevates pruning from a local snapshot decision to a temporal trajectory modeling problem.<n>This enables a dynamic rectification mechanism that selectively reactivates "late-blooming" tokens, those initially undervalued but exhibiting rising semantic importance, thereby preventing the loss of critical visual cues.
  arXiv  Detail & Related papers  (2026-07-30T15:07:00Z)
• ERA: Entropy-Guided Visual Token Pruning with Rectified Attention for Efficient MLLMs [77.57757084153083]
  We propose ERA, an Entropy-guided visual token pruning framework with Rectified Attention for efficient MLLMs.<n> ERA comprises three crucial components: Dual-view Entropy Pruning (DEP), Bias-aware Token Recycling (BTR), and Logit-preserving Attention Rectification (LAR)
  arXiv  Detail & Related papers  (2026-06-30T17:20:29Z)
• One Layer's Trash is Another Layer's Treasure: Adaptive Layer-wise Visual Token Selection in LVLMs [18.48496973561215]
  Large Vision-Language Models (LVLMs) have achieved remarkable success across diverse multimodal tasks, yet their practical deployment remains constrained by the computational burden arising from lengthy visual tokens.<n>We propose Adaptive Layer-wise Visual Token Selection (ALVTS), a novel framework that breaks away from the conventional static token pruning paradigm.<n>With an 89% token compression ratio, ALVTS retains 96.7% of the original model's accuracy, achieving a superior efficiency-accuracy trade-off for LVLM inference.
  arXiv  Detail & Related papers  (2026-06-12T08:58:58Z)
• Look Less, Reason More: Block-wise Attention Skipping for Efficient Multimodal LLMs [89.7106332677868]
  We present Visual-Skip, a training-free inference paradigm that decouples spatial interaction from semantic evolution.<n>V-Skip effectively bypasses redundant vision attention to achieve block-wise sparsity, maintaining a 94.16% to 100.31% performance retention across diverse MLLMs.
  arXiv  Detail & Related papers  (2026-06-07T08:32:13Z)
• Q Cache: Visual Attention is Valuable in Less than Half of Decode Layers for Multimodal Large Language Model [21.206033754351786]
  Multimodal large language models (MLLMs) are plagued by exorbitant inference costs attributable to the profusion of visual tokens.<n>Existing approaches focus on token-wise optimization, leveraging diverse token pruning techniques to eliminate non-crucial visual tokens.<n>We propose Lazy Attention, an efficient attention mechanism that enables cross-layer sharing of similar attention patterns.
  arXiv  Detail & Related papers  (2026-02-02T10:08:00Z)
• Don't Just Chase "Highlighted Tokens" in MLLMs: Revisiting Visual Holistic Context Retention [50.97683288777336]
  Multimodal Large Language Models (MLLMs) suffer from considerable computational overhead due to their reliance on massive visual tokens.<n>Recent studies have explored token pruning to alleviate this problem, which typically uses text-vision cross-attention.<n>We propose HoloV, a plug-and-play visual token pruning framework for efficient inference.
  arXiv  Detail & Related papers  (2025-10-03T11:33:40Z)
• Variation-aware Vision Token Dropping for Faster Large Vision-Language Models [24.952668143243542]
  Large vision-language models (LVLMs) have demonstrated remarkable capabilities in multimodal understanding tasks.<n> Token compression offers a direct solution by reducing the number of tokens to be processed, thereby improving computational efficiency.<n>We propose Variation-aware Vision Token Dropping (textiti.e., textbfV$2$Drop), which progressively removes visual tokens with minimal variation during LVLM inference.
  arXiv  Detail & Related papers  (2025-09-01T15:28:44Z)
• Rethinking Visual Token Reduction in LVLMs under Cross-modal Misalignment [38.04426918886084]
  Vision-Language Models (LVLMs) encode visual inputs as dense sequences of patch-level tokens to capture fine-grained semantics.<n>Previous efforts have explored visual token reduction either prior to or within the large language models (LLMs)<n>We introduce VisionDrop, a training-free, visual-only pruning framework that selects informative visual tokens based on intra-modal (visual-to-visual) attention.
  arXiv  Detail & Related papers  (2025-06-27T14:55:40Z)
• ToDRE: Visual Token Pruning via Diversity and Task Awareness for Efficient Large Vision-Language Models [59.47738955960352]
  ToDRE is a two-stage and training-free token compression framework.<n>It achieves superior performance by pruning tokens based on token Diversity and token-task RElevance.
  arXiv  Detail & Related papers  (2025-05-24T15:47:49Z)
• Efficient LLaMA-3.2-Vision by Trimming Cross-attended Visual Features [24.33252753245426]
  We exploit the sparse nature in cross-attention maps to selectively prune redundant visual features.<n>Our model can reduce inference latency and memory usage while achieving benchmark parity.
  arXiv  Detail & Related papers  (2025-04-01T09:10:32Z)
• TopV: Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision Language Model [56.43860351559185]
  We introduce textbfTopV, a compatible textbfTOken textbfPruning with inference Time Optimization for fast and low-memory textbfVLM.<n>Our framework incorporates a visual-aware cost function to measure the importance of each source visual token, enabling effective pruning of low-importance tokens.
  arXiv  Detail & Related papers  (2025-03-24T01:47:26Z)
• RedundancyLens: Revealing and Exploiting Visual Token Processing Redundancy for Efficient Decoder-Only MLLMs [38.34856927170692]
  We propose a training-free framework for analyzing trained Multimodal Large Language Model (MLLM)<n>It consists of Probe-Activated Dynamic FFN and Hollow Attention, which enable adjustable reductions in computations for visual tokens.<n>Experiments demonstrate substantial, structured, and clustered redundancy unique to decoder-only MLLMs.
  arXiv  Detail & Related papers  (2025-01-31T11:09:16Z)
• Inference Optimal VLMs Need Fewer Visual Tokens and More Parameters [54.01228554126122]
  Vision Language Models (VLMs) have demonstrated strong capabilities across various visual understanding and reasoning tasks.<n>To reduce inference costs, one can either downsize the Large Language Models (LLMs) or reduce the number of input tokens needed to represent the image.<n>We take the first steps toward designing token compression algorithms tailored for high-compression settings.
  arXiv  Detail & Related papers  (2024-11-05T18:54:21Z)

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.