Cover Image

Focusing by Contrastive Attention: Enhancing VLMs' Visual Reasoning

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Ge, Yuyao, Liu, Shenghua, Wang, Yiwei, Mei, Lingrui, Bi, Baolong, Zhou, Xuanshan, Yao, Jiayu, Guo, Jiafeng, Cheng, Xueqi
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

HighlightBench: Benchmarking Markup-Driven Table Reasoning in Scientific Documents
by: Wang, Lexin, et al.
Published: (2026)

Innate Reasoning is Not Enough: In-Context Learning Enhances Reasoning Large Language Models with Less Overthinking
by: Ge, Yuyao, et al.
Published: (2025)

Prism-$Δ$: Differential Subspace Steering for Prompt Highlighting in Large Language Models
by: Ge, Yuyao, et al.
Published: (2026)

Reward and Guidance through Rubrics: Promoting Exploration to Improve Multi-Domain Reasoning
by: Bi, Baolong, et al.
Published: (2025)

Decoding by Contrasting Knowledge: Enhancing LLMs' Confidence on Edited Facts
by: Bi, Baolong, et al.
Published: (2024)

Who is in the Spotlight: The Hidden Bias Undermining Multimodal Retrieval-Augmented Generation
by: Yao, Jiayu, et al.
Published: (2025)

Gated Differentiable Working Memory for Long-Context Language Modeling
by: Mei, Lingrui, et al.
Published: (2026)

SLANG: New Concept Comprehension of Large Language Models
by: Mei, Lingrui, et al.
Published: (2024)

LPNL: Scalable Link Prediction with Large Language Models
by: Bi, Baolong, et al.
Published: (2024)

Can Graph Descriptive Order Affect Solving Graph Problems with LLMs?
by: Ge, Yuyao, et al.
Published: (2024)

a1: Steep Test-time Scaling Law via Environment Augmented Generation
by: Mei, Lingrui, et al.
Published: (2025)

Not in Sync: Unveiling Temporal Bias in Audio Chat Models
by: Yao, Jiayu, et al.
Published: (2025)

"Not Aligned" is Not "Malicious": Being Careful about Hallucinations of Large Language Models' Jailbreak
by: Mei, Lingrui, et al.
Published: (2024)

HiddenGuard: Fine-Grained Safe Generation with Specialized Representation Router
by: Mei, Lingrui, et al.
Published: (2024)

MVAM: Multi-View Attention Method for Fine-grained Image-Text Matching
by: Cui, Wanqing, et al.
Published: (2024)

Parameters vs. Context: Fine-Grained Control of Knowledge Reliance in Language Models
by: Bi, Baolong, et al.
Published: (2025)

Adaptive Token Biaser: Knowledge Editing via Biasing Key Entities
by: Bi, Baolong, et al.
Published: (2024)

StruEdit: Structured Outputs Enable the Fast and Accurate Knowledge Editing for Large Language Models
by: Bi, Baolong, et al.
Published: (2024)

A Survey of Vibe Coding with Large Language Models
by: Ge, Yuyao, et al.
Published: (2025)

PromptCD: Test-Time Behavior Enhancement via Polarity-Prompt Contrastive Decoding
by: Bi, Baolong, et al.
Published: (2026)

RefineX: Learning to Refine Pre-training Data at Scale from Expert-Guided Programs
by: Bi, Baolong, et al.
Published: (2025)

CLIPure: Purification in Latent Space via CLIP for Adversarially Robust Zero-Shot Classification
by: Zhang, Mingkun, et al.
Published: (2025)

Visual Transformation Telling
by: Cui, Wanqing, et al.
Published: (2023)

TIR-Flow: Active Video Search and Reasoning with Frozen VLMs
by: Jin, Hongbo, et al.
Published: (2026)

Classifier Guidance Enhances Diffusion-based Adversarial Purification by Preserving Predictive Information
by: Zhang, Mingkun, et al.
Published: (2024)

CausalDiff: Causality-Inspired Disentanglement via Diffusion Model for Adversarial Defense
by: Zhang, Mingkun, et al.
Published: (2024)

Is Factuality Enhancement a Free Lunch For LLMs? Better Factuality Can Lead to Worse Context-Faithfulness
by: Bi, Baolong, et al.
Published: (2024)

Adaptive Chain-of-Focus Reasoning via Dynamic Visual Search and Zooming for Efficient VLMs
by: Zhang, Xintong, et al.
Published: (2025)

VisualActBench: Can VLMs See and Act like a Human?
by: Zhang, Daoan, et al.
Published: (2025)

Through the Lens of Contrast: Self-Improving Visual Reasoning in VLMs
by: Pan, Zhiyu, et al.
Published: (2026)

Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs
by: Zhang, Qizhe, et al.
Published: (2024)

Vision Verification Enhanced Fusion of VLMs for Efficient Visual Reasoning
by: Tekin, Selim Furkan, et al.
Published: (2026)

A Survey of Context Engineering for Large Language Models
by: Mei, Lingrui, et al.
Published: (2025)

Thinking with Gaze: Sequential Eye-Tracking as Visual Reasoning Supervision for Medical VLMs
by: Li, Yiwei, et al.
Published: (2026)

ViFP: A Framework for Visual False Positive Detection to Enhance Reasoning Reliability in VLMs
by: Zhang, Ben, et al.
Published: (2025)

Beyond Shortcuts: Mitigating Visual Illusions in Frozen VLMs via Qualitative Reasoning
by: Guo, Hao, et al.
Published: (2026)

VLMs have Tunnel Vision: Evaluating Nonlocal Visual Reasoning in Leading VLMs
by: Berman, Shmuel, et al.
Published: (2025)

Towards Fully Exploiting LLM Internal States to Enhance Knowledge Boundary Perception
by: Ni, Shiyu, et al.
Published: (2025)

SemVink: Advancing VLMs' Semantic Understanding of Optical Illusions via Visual Global Thinking
by: Li, Sifan, et al.
Published: (2025)

Decomposing Visual Classification: Assessing Tree-Based Reasoning in VLMs
by: Elmansoury, Sary, et al.
Published: (2025)