arXiv:2608. 07193v1 Announce Type: new Abstract: Visual-token pruning can substantially reduce the inference cost of multimodal large language models (MLLMs), yet existing methods largely rely on fixed, handcrafted heuristics and costly expert trial and error.
arXiv:2606. 27161v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have achieved strong multimodal reasoning capabilities, but their efficiency is limited by the large number of visual tokens, which introduces substantial computational overhead.
The paper shows that only a small subset of attention heads in vision-language models is responsible for selecting critical visual tokens. By pruning tokens based on similarity before LLM reasoning and then applying head‑aware pruning during reasoning, the proposed ProViP framework achieves high task performance with significant speedups. Experiments on LLaVA‑1.5‑7B demonstrate 95.9% performance retention and a 1.62× inference speedup at an 88.9% pruning ratio.
arXiv:2609.10346v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) process hundreds or thousands of visual tokens per image, incurring prohibitive inference costs. While existin...
SinkPruner is a training‑free framework that prunes visual tokens for multimodal large language models by first removing high‑norm redundant tokens with a visual sanitizer and then selectively keeping tokens that align with the text query using a text‑guided pruner. The coarse‑to‑fine design reduces attention sink and dispersion, enabling an 89% token reduction while preserving 96.5% of LLaVA‑1.5’s performance and 91.8% of Qwen2.5‑VL’s performance across twelve image‑language and four video‑language benchmarks. The visual sanitizer also improves existing pruning methods, showing strong transferability.
arXiv:2609.05916v1 Announce Type: cross Abstract: Large vision-language models (LVLMs) achieve strong multimodal understanding, but the hundreds to thousands of visual tokens they process impose subs...
arXiv:2609.34330v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) have demonstrated impressive performance in multimodal understanding, but processing large numbers of visu...
Multimodal foundation models are reshaping edge-cloud visual intelligence from task-specific feature pipelines into token-based interfaces, where edge devices encode visual inputs into tokens for a general-purpose cloud MLLM. However, dense visual-token sequences increase cloud-side inference costs.
The paper argues that token importance alone is insufficient to determine safe removal of visual tokens in multimodal large language models, because removability depends on representation depth and the surrounding deletion set. Through controlled experiments, the authors show that the same tokens can have different effects when removed at different depths or contexts. They introduce CoRePrune, a training‑free two‑stage pruning framework that refreshes deletion effects as visual representations evolve and refines candidate tokens based on the current deletion set, achieving high performance retention across multiple backbones and reducing prefill time significantly.
arXiv:2609.13804v1 Announce Type: new Abstract: Visual prefixes account for a major portion of the per-layer computation in multimodal large language models (MLLMs), making visual-token pruning a dir...
VisCache introduces a two-stage, plug‑and‑play framework for pruning visual key‑value caches in Vision Large Language Models without retraining. The first stage filters out temporally redundant keyframes, while the second stage, PruneKV, applies a parabolic layer‑wise budget and asymmetric update to selectively prune keys and fuse values, preserving essential context. Experiments show up to 2.35× speedup and significant memory savings with only 19–28% of the original cache retained, outperforming existing baselines.
arXiv:2609.15131v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) require substantial computation to process numerous visual tokens across all transformer layers. Most method...
The paper introduces Adaptive Visual Token Pruning (AVTP), a training‑free framework that dynamically selects pruning layers and ratios for large vision‑language models (LVLMs) when processing multiple image sequences. By analyzing visual attention distributions across different LVLM architectures, AVTP adapts token retention to image importance, enabling efficient inference without relying on attention‑based computations incompatible with FlashAttention. Experiments show significant speedups—up to 2× for Qwen3VL‑8B—while preserving or even improving accuracy on multi‑image benchmarks.