Cover Image

To Sink or Not to Sink: Visual Information Pathways in Large Vision-Language Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Luo, Jiayun, Fan, Wan-Cyuan, Wang, Lyuyang, He, Xiangteng, Rahman, Tanzila, Abolmaesumi, Purang, Sigal, Leonid
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition Artificial Intelligence Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

MMFactory: A Universal Solution Search Engine for Vision-Language Tasks
by: Fan, Wan-Cyuan, et al.
Published: (2024)

Tinted Frames: Question Framing Blinds Vision-Language Models
by: Fan, Wan-Cyuan, et al.
Published: (2026)

Response Wide Shut: Surprising Observations in Basic Vision Language Model Capabilities
by: Chandhok, Shivam, et al.
Published: (2024)

On Pre-training of Multimodal Language Models Customized for Chart Understanding
by: Fan, Wan-Cyuan, et al.
Published: (2024)

In-Depth and In-Breadth: Pre-training Multimodal Language Models Customized for Comprehensive Chart Understanding
by: Fan, Wan-Cyuan, et al.
Published: (2025)

All in One: A Unified Synthetic Data Pipeline for Multimodal Video Understanding
by: Rahman, Tanzila, et al.
Published: (2026)

The ART of Composition: Attention-Regularized Training for Compositional Visual Grounding
by: Luo, Jiayun, et al.
Published: (2024)

When Sinks Help or Hurt: Unified Framework for Attention Sink in Large Vision-Language Models
by: Choi, Jiho, et al.
Published: (2026)

TAM-VT: Transformation-Aware Multi-scale Video Transformer for Segmentation and Tracking
by: Goyal, Raghav, et al.
Published: (2023)

ChartGaze: Enhancing Chart Understanding in LVLMs with Eye-Tracking Guided Attention Refinement
by: Salamatian, Ali, et al.
Published: (2025)

Emergent Open-Vocabulary Semantic Segmentation from Off-the-shelf Vision-Language Models
by: Luo, Jiayun, et al.
Published: (2023)

SinkTrack: Attention Sink based Context Anchoring for Large Language Models
by: Liu, Xu, et al.
Published: (2026)

SCALE-VLP: Soft-Weighted Contrastive Volumetric Vision-Language Pre-training with Spatial-Knowledge Semantics
by: Mahdizadeh, Ailar, et al.
Published: (2025)

Visual Concept-driven Image Generation with Text-to-Image Diffusion Model
by: Rahman, Tanzila, et al.
Published: (2024)

InvAD: Inversion-based Reconstruction-Free Anomaly Detection with Diffusion Models
by: Sakai, Shunsuke, et al.
Published: (2025)

CoRDS: Coreset-based Representative and Diverse Selection for Streaming Video Understanding
by: Mahdizadeh, Ailar, et al.
Published: (2026)

EchoAgent: Guideline-Centric Reasoning Agent for Echocardiography Measurement and Interpretation
by: Daghyani, Matin, et al.
Published: (2025)

Rethinking Visual Dependency in Long-Context Reasoning for Large Vision-Language Models
by: Zhou, Yucheng, et al.
Published: (2024)

Artifacts and Attention Sinks: Structured Approximations for Efficient Vision Transformers
by: Lu, Andrew, et al.
Published: (2025)

DySink: Dynamic Frame Sinks for Autoregressive Long Video Generation
by: Ye, Bo, et al.
Published: (2026)

Garbage Attention in Large Language Models: BOS Sink Heads and Sink-aware Pruning
by: Sok, Jaewon, et al.
Published: (2026)

Visual In-Context Learning for Large Vision-Language Models
by: Zhou, Yucheng, et al.
Published: (2024)

See What You Are Told: Visual Attention Sink in Large Multimodal Models
by: Kang, Seil, et al.
Published: (2025)

CTR-Sink: Attention Sink for Language Models in Click-Through Rate Prediction
by: Li, Zixuan, et al.
Published: (2025)

SPIKE-RL: Video-LLMs meet Bayesian Surprise
by: Ravi, Sahithya, et al.
Published: (2025)

VisOnlyQA: Large Vision Language Models Still Struggle with Visual Perception of Geometric Information
by: Kamoi, Ryo, et al.
Published: (2024)

ProtoTTA: Prototype-Guided Test-Time Adaptation
by: Abootorabi, Mohammad Mahdi, et al.
Published: (2026)

CCSI: Continual Class-Specific Impression for Data-free Class Incremental Learning
by: Ayromlou, Sana, et al.
Published: (2024)

Pseudo-D: Informing Multi-View Uncertainty Estimation with Calibrated Neural Training Dynamics
by: Gu, Ang Nan, et al.
Published: (2025)

Large Vision-Language Models for Remote Sensing Visual Question Answering
by: Siripong, Surasakdi, et al.
Published: (2024)

BabyVision: Visual Reasoning Beyond Language
by: Chen, Liang, et al.
Published: (2026)

A Comprehensive Information-Decomposition Analysis of Large Vision-Language Models
by: Xiu, Lixin, et al.
Published: (2026)

VividMed: Vision Language Model with Versatile Visual Grounding for Medicine
by: Luo, Lingxiao, et al.
Published: (2024)

PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction
by: Xing, Long, et al.
Published: (2024)

DSeq-JEPA: Discriminative Sequential Joint-Embedding Predictive Architecture
by: He, Xiangteng, et al.
Published: (2025)

Judging the Judges: Can Large Vision-Language Models Fairly Evaluate Chart Comprehension and Reasoning?
by: Laskar, Md Tahmid Rahman, et al.
Published: (2025)

A Comprehensive Analysis for Visual Object Hallucination in Large Vision-Language Models
by: Jing, Liqiang, et al.
Published: (2025)

Instruction-Aligned Visual Attention for Mitigating Hallucinations in Large Vision-Language Models
by: Li, Bin, et al.
Published: (2025)

VScan: Rethinking Visual Token Reduction for Efficient Large Vision-Language Models
by: Zhang, Ce, et al.
Published: (2025)

ChartMuseum: Testing Visual Reasoning Capabilities of Large Vision-Language Models
by: Tang, Liyan, et al.
Published: (2025)