arXiv Computation and Language

Semantic Head Specialization Guides Hybrid ViT Attention for Multimodal LLMs

The paper introduces Semantic Head Specialization (SHS), a phenomenon where Vision Transformer (ViT) attention heads specialize as either object- or background-focused, most evident under full attention. It proposes the SHS-Index to quantify this specialization, demonstrating its ability to distinguish full-attention from chunk-window ViTs and its strong correlation with downstream benchmark performance. Leveraging insights into window interaction, token serialization, and local softmax allocation, the authors design Ariadne Attention, a hybrid attention mechanism that matches full-attention performance on 22 image and video tasks while reducing attention compute by 6.5×.

arXiv Computer Vision

Aug 27

RAVE: Re-Allocating Visual Attention in Large Multimodal Models

RAVE (Re-Allocating Visual Attention) is a lightweight pair‑gating mechanism that adds a learned query‑key bias to pre‑softmax attention scores over visual keys, derived from pre‑RoPE query and key features. It requires no architectural changes to the backbone and can be trained end‑to‑end with the rest of the model. Across multiple multimodal benchmarks, RAVE improves standard attention by an average of 3 points, especially on perception‑intensive tasks such as multilingual OCR, chart understanding, document VQA, and scene text VQA.

By Xi Leng, Xinhong Ma, Ziqiang Dong, Feng Zhang, Xiaoying Tang, Yang Yang, Guanjun Jiang

computer-vision multimodal benchmarks safety

arXiv Computer Vision

Aug 26

HAP: Head-Adaptive Visual Token Pruning via Cross-Modal Alignment

arXiv:2608.23921v1 Announce Type: new Abstract: Recent Vision-Language Models encode high-resolution images into long visual token sequences, incurring prohibitive prefill costs. To compress them, ex...

By Yuanhao Sun, Huawei Ji, Yuan Jin, Cheng Deng, Luoyi Fu, Xinbing Wang

llms efficiency multimodal benchmarks safety

arXiv AI

Jun 6

Mechanistic Insights into Functional Sparsity in Multimodal LLMs via CoRe Heads

arXiv:2606. 05843v1 Announce Type: cross Abstract: While Multimodal Large Language Models (MLLMs) demonstrate remarkable proficiency on complex vision-language tasks, the mechanisms by which they extract query-relevant visual features from complex, noisy contexts remain opaque.

By Ruoxi Sun, Quantong Qiu, Juntao Li, Zecheng Tang, Yihang Lou, Min Zhang

llms efficiency multimodal safety

arXiv Machine Learning

Jun 16

Rethinking the Role of Efficient Attention in Hybrid Architectures

arXiv:2606. 15378v1 Announce Type: cross Abstract: Modern language models increasingly adopt hybrid architectures that combine full attention with efficient attention modules, such as sliding-window attention (SWA) and recurrent sequence mixers.

By Ziqing Qiao, Yinuo Xu, Chaojun Xiao, Zhou Su, Zihan Zhou, Yingfa Chen, Xiaoyue Xu, Xu Han, Zhiyuan Liu

llms

arXiv Computer Vision

Aug 27

Not All Attention Heads Contribute to Critical Visual Token Selection: Head-Aware Pruning Matters More

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.

By Chaofang Ma, Lin Jiang, Carol Jingyi Li, Xingyu Liu, Zeyu Li, Jiang Xu, Wei Zhang

llms rag efficiency multimodal

arXiv AI

Jun 12

HYDRA-X: Native Unified Multimodal Models with Holistic Visual Tokenizers

arXiv:2606. 13289v1 Announce Type: cross Abstract: Holistic visual tokenizers are fundamental to unified multimodal models (UMMs) as they map diverse visual inputs into a unified representation space.

By Guozhen Zhang, Xuerui Qiu, Yutao Cui, Tianhui Song, Changlin Li, Junzhe Li, Tao Huang, Xiao Zhang, Yang Li, Jianbing Wu, Miles Yang, Zhao Zhong, Liefeng Bo, Limin Wang

llms rag diffusion computer-vision nlp multimodal

arXiv Machine Learning

Jun 5

GRAMformer: Any-Order Modality Interactions via Volumetric Multimodal Cross-Attention

arXiv:2606. 06249v1 Announce Type: cross Abstract: Transformer-based multimodal models rely on attention mechanisms to integrate information across heterogeneous modalities.

By Giordano Cicchetti, Eleonora Grassucci, Danilo Comminiello

llms multimodal

arXiv AI

Aug 10

Learning to Predict Middle-Layer Attention in MLLMs for Visual Token Prunin

arXiv:2608. 06411v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) achieve strong performance across diverse vision-language tasks, but their efficiency is limited by the cost of processing numerous visual tokens.

By Yuyao Sun, Tao Deng, Shuang Li, Deqing Wang, Hao Geng, Minjun Yu

llms efficiency multimodal benchmarks

arXiv AI

Jul 17

VideoSEMA: a scalable and efficient Mamba-like attention for video understanding

arXiv:2607. 14711v1 Announce Type: cross Abstract: We present for video understanding (classification) a split space-time attention model, VideoSEMA, consisting of a scalable and efficient Mamba-like attention (SEMA) block in space and a softmax temporal attention in time.

By Nhat Thanh Tran, Fanghui Xue andShuai Zhang, Jiancheng Lyu, Yunling Zheng, Yingyong Qi, Jack Xin

llms computer-vision fine-tuning benchmarks

arXiv Machine Learning

Jun 29

Prism Transformer: Progressive Head Schedules for Hierarchical Attention Processing

arXiv:2606. 27449v1 Announce Type: new Abstract: Multi-head attention conventionally partitions the hidden dimension equally across all heads at every layer, enforcing an identical representational subspace dimension (dh = dmodel/h) throughout the models depth.

By Shubham Aggarwal

llms benchmarks

arXiv AI

Jul 2

Stateful Token Reduction for Long-Video Hybrid VLMs

arXiv:2603. 00198v2 Announce Type: replace-cross Abstract: Token reduction accelerates long-video vision--language models (VLMs), but existing methods target Transformers, where reduction is treated as token pruning.

By Jindong Jiang, Amala Sanjay Deshmukh, Kateryna Chumachenko, Karan Sapra, Zhiding Yu, Guilin Liu, Andrew Tao, Pavlo Molchanov, Jan Kautz, Wonmin Byeon

llms fine-tuning efficiency benchmarks

arXiv Computer Vision

3d ago

Enhancing Self-Supervised Visual Representation Learning via Low-Rank Adapted LLMs

arXiv:2507.00754v3 Announce Type: replace Abstract: The integration of Large Language Model (LLMs) blocks with Vision Transformers (ViTs) holds immense promise for vision-only tasks by leveraging the...

By Selim Kuzucu, Muhammad Ferjad Naeem, Anna Kukleva, Federico Tombari, Bernt Schiele

llms fine-tuning

Related stories