Imagen de Portada

BlackMirror: Black-Box Backdoor Detection for Text-to-Image Models via Instruction-Response Deviation

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Li, Feiran, Xu, Qianqian, Bao, Shilong, Yang, Zhiyong, Zhao, Xilin, Cao, Xiaochun, Huang, Qingming
Formato:	Preprint
Publicado:	2026
Materias:	Computer Vision and Pattern Recognition Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

One Image is Worth a Thousand Words: A Usability Preservable Text-Image Collaborative Erasing Framework
por: Li, Feiran, et al.
Publicado: (2025)

Towards Size-invariant Salient Object Detection: A Generic Evaluation and Optimization Approach
por: Bao, Shilong, et al.
Publicado: (2025)

Size-invariance Matters: Rethinking Metrics and Losses for Imbalanced Multi-object Salient Object Detection
por: Li, Feiran, et al.
Publicado: (2024)

MixBridge: Heterogeneous Image-to-Image Backdoor Attack through Mixture of Schrödinger Bridges
por: Qin, Shixi, et al.
Publicado: (2025)

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

Hybrid Generative Fusion for Efficient and Privacy-Preserving Face Recognition Dataset Generation
por: Li, Feiran, et al.
Publicado: (2025)

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

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

LightFair: Towards an Efficient Alternative for Fair T2I Diffusion via Debiasing Pre-trained Text Encoders
por: Han, Boyu, et al.
Publicado: (2025)

Bidirectional Logits Tree: Pursuing Granularity Reconcilement in Fine-Grained Classification
por: Lu, Zhiguang, et al.
Publicado: (2024)

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

Understanding-Enhanced Model Collaboration for Long-Tailed Egocentric Mistake Detection
por: Han, Boyu, et al.
Publicado: (2026)

Dual-Stage Reweighted MoE for Long-Tailed Egocentric Mistake Detection
por: Han, Boyu, et al.
Publicado: (2025)

Suppress Content Shift: Better Diffusion Features via Off-the-Shelf Generation Techniques
por: Meng, Benyuan, et al.
Publicado: (2024)

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

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

DirMixE: Harnessing Test Agnostic Long-tail Recognition with Hierarchical Label Vartiations
por: Yang, Zhiyong, et al.
Publicado: (2024)

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

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

From Static to Dynamic: Exploring Self-supervised Image-to-Video Representation Transfer Learning
por: Liu, Yang, et al.
Publicado: (2026)

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

Top-K Pairwise Ranking: Bridging the Gap Among Ranking-Based Measures for Multi-Label Classification
por: Wang, Zitai, et al.
Publicado: (2024)

Prompting the Unseen: Detecting Hidden Backdoors in Black-Box Models
por: Huang, Zi-Xuan, et al.
Publicado: (2024)

Mind the Way You Select Negative Texts: Pursuing the Distance Consistency in OOD Detection with VLMs
por: Xu, Zhikang, et al.
Publicado: (2026)

Not All Diffusion Model Activations Have Been Evaluated as Discriminative Features
por: Meng, Benyuan, et al.
Publicado: (2024)

Dynamic Attention Analysis for Backdoor Detection in Text-to-Image Diffusion Models
por: Wang, Zhongqi, et al.
Publicado: (2025)

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

Low-Frequency Black-Box Backdoor Attack via Evolutionary Algorithm
por: Qiao, Yanqi, et al.
Publicado: (2024)

BDFirewall: Towards Effective and Expeditiously Black-Box Backdoor Defense in MLaaS
por: Li, Ye, et al.
Publicado: (2025)

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

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

Robust Anti-Backdoor Instruction Tuning in LVLMs
por: Xun, Yuan, et al.
Publicado: (2025)

Quantifying the Potential to Escape Filter Bubbles: A Behavior-Aware Measure via Contrastive Simulation
por: Feng, Difu, et al.
Publicado: (2025)

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

PersGuard: Preventing Malicious Personalization via Backdoor Attacks on Pre-trained Text-to-Image Diffusion Models
por: Liu, Xinwei, et al.
Publicado: (2025)

T2IShield: Defending Against Backdoors on Text-to-Image Diffusion Models
por: Wang, Zhongqi, et al.
Publicado: (2024)

Sequential Manipulation Against Rank Aggregation: Theory and Algorithm
por: Ma, Ke, et al.
Publicado: (2024)

Test-time Backdoor Mitigation for Black-Box Large Language Models with Defensive Demonstrations
por: Mo, Wenjie, et al.
Publicado: (2023)

Illuminating the Black Box: Real-Time Monitoring of Backdoor Unlearning in CNNs via Explainable AI
por: Hoang, Tien Dat
Publicado: (2025)

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