Cover Image

Localization then Neutralization: Gradient-guided Token Suppression against Visual Prompt Injection Attack

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhang, Dongpeng, Ma, Ke, Jiang, Yangbangyan, Pei, Gaozheng, Huang, Longtao, Xu, Qianqian, Huang, Qingming
Format:	Preprint
Published:	2026
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

A Unified Framework for Stealthy Adversarial Generation via Latent Optimization and Transferability Enhancement
by: Pei, Gaozheng, et al.
Published: (2025)

Regularized Contrastive Partial Multi-view Outlier Detection
by: Wang, Yijia, et al.
Published: (2024)

Diffusion-based Adversarial Purification from the Perspective of the Frequency Domain
by: Pei, Gaozheng, et al.
Published: (2025)

HGOE: Hybrid External and Internal Graph Outlier Exposure for Graph Out-of-Distribution Detection
by: He, Junwei, et al.
Published: (2024)

Closing the Approximation Gap of Partial AUC Optimization: A Tale of Two Formulations
by: Jiang, Yangbangyan, et al.
Published: (2025)

AUCSeg: AUC-oriented Pixel-level Long-tail Semantic Segmentation
by: Han, Boyu, et al.
Published: (2024)

Enhancing Sample Utilization in Noise-Robust Deep Metric Learning With Subgroup-Based Positive-Pair Selection
by: Yu, Zhipeng, et al.
Published: (2025)

Divide and Conquer: Heterogeneous Noise Integration for Diffusion-based Adversarial Purification
by: Pei, Gaozheng, et al.
Published: (2025)

Exploring Query Efficient Data Generation towards Data-free Model Stealing in Hard Label Setting
by: Pei, Gaozheng, et al.
Published: (2024)

OpenworldAUC: Towards Unified Evaluation and Optimization for Open-world Prompt Tuning
by: Hua, Cong, et al.
Published: (2025)

The Bridge-Garden Dilemma in LLM Distillation: Why Mixing Hard and Soft Labels Works
by: Wang, Guanghui, et al.
Published: (2026)

A Unified Perspective for Loss-Oriented Imbalanced Learning via Localization
by: Wang, Zitai, et al.
Published: (2023)

Making Training-Free Diffusion Segmentors Scale with the Generative Power
by: Meng, Benyuan, et al.
Published: (2026)

SSE-SAM: Balancing Head and Tail Classes Gradually through Stage-Wise SAM
by: Lyu, Xingyu, et al.
Published: (2024)

IterInject: Indirect Prompt Injection Against LLM Agents via Feedback-Guided Iterative Optimization
by: Chen, Zixuan, et al.
Published: (2026)

Harmless Yet Harmful: Neutral Prompting Attacks for Stealthy Hallucination Steering in Agent Skills
by: Hsu, Chia-Yi, et al.
Published: (2026)

The Devil is in the Condition Numbers: Why is GLU Better than non-GLU Structure?
by: Lyu, Xingyu, et al.
Published: (2026)

Guiding Diffusion-based Reconstruction with Contrastive Signals for Balanced Visual Representation
by: Han, Boyu, et al.
Published: (2026)

Semantic Concentration for Self-Supervised Dense Representations Learning
by: Wen, Peisong, et al.
Published: (2025)

ABKD: Pursuing a Proper Allocation of the Probability Mass in Knowledge Distillation via $α$-$β$-Divergence
by: Wang, Guanghui, et al.
Published: (2025)

ReconBoost: Boosting Can Achieve Modality Reconcilement
by: Hua, Cong, et al.
Published: (2024)

Collective Certified Robustness against Graph Injection Attacks
by: Lai, Yuni, et al.
Published: (2024)

PINA: Prompt Injection Attack against Navigation Agents
by: Liu, Jiani, et al.
Published: (2026)

Prompt Injection Attacks on LLM Generated Reviews of Scientific Publications
by: Keuper, Janis
Published: (2025)

Formalizing and Benchmarking Prompt Injection Attacks and Defenses
by: Liu, Yupei, et al.
Published: (2023)

Focal-SAM: Focal Sharpness-Aware Minimization for Long-Tailed Classification
by: Li, Sicong, et al.
Published: (2025)

Improved Diversity-Promoting Collaborative Metric Learning for Recommendation
by: Bao, Shilong, et al.
Published: (2024)

TuckA: Hierarchical Compact Tensor Experts for Efficient Fine-Tuning
by: Lei, Qifeng, et al.
Published: (2025)

Optimizing Partial Area Under the Top-k Curve: Theory and Practice
by: Wang, Zitai, et al.
Published: (2022)

Measuring Real-World Prompt Injection Attacks in LLM-based Resume Screening
by: Zhang, Mohan, et al.
Published: (2026)

Bypassing LLM Guardrails: An Empirical Analysis of Evasion Attacks against Prompt Injection and Jailbreak Detection Systems
by: Hackett, William, et al.
Published: (2025)

Prompt Injection Attacks on Large Language Models in Oncology
by: Clusmann, Jan, et al.
Published: (2024)

Defending Against Indirect Prompt Injection Attacks With Spotlighting
by: Hines, Keegan, et al.
Published: (2024)

Attention Tracker: Detecting Prompt Injection Attacks in LLMs
by: Hung, Kuo-Han, et al.
Published: (2024)

Design Patterns for Securing LLM Agents against Prompt Injections
by: Beurer-Kellner, Luca, et al.
Published: (2025)

Empirical Analysis of Large Vision-Language Models against Goal Hijacking via Visual Prompt Injection
by: Kimura, Subaru, et al.
Published: (2024)

Exploring Potential Prompt Injection Attacks in Federated Military LLMs and Their Mitigation
by: Lee, Youngjoon, et al.
Published: (2025)

Cross-Context Backdoor Attacks against Graph Prompt Learning
by: Lyu, Xiaoting, et al.
Published: (2024)

Adaptive Attacks Break Defenses Against Indirect Prompt Injection Attacks on LLM Agents
by: Zhan, Qiusi, et al.
Published: (2025)

Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attacks
by: Xue, Lulu, et al.
Published: (2024)