Imagen de Portada

Trusted Weights, Treacherous Optimizations? Optimization-Triggered Backdoor Attacks on LLMs

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Wang, Yifei, Li, Tianlin, Zhang, Xiaohan, Yang, Yida, Zhang, Xiaoyu, Pan, Li
Formato:	Preprint
Publicado:	2026
Materias:	Cryptography and Security Artificial Intelligence Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Compromising Embodied Agents with Contextual Backdoor Attacks
por: Liu, Aishan, et al.
Publicado: (2024)

Concealing Backdoor Model Updates in Federated Learning by Trigger-Optimized Data Poisoning
por: Zhang, Yujie, et al.
Publicado: (2024)

Revisiting Backdoor Attacks on Time Series Classification in the Frequency Domain
por: Huang, Yuanmin, et al.
Publicado: (2025)

The Art of Deception: Robust Backdoor Attack using Dynamic Stacking of Triggers
por: Mengara, Orson
Publicado: (2024)

BadAgent: Inserting and Activating Backdoor Attacks in LLM Agents
por: Wang, Yifei, et al.
Publicado: (2024)

Rethinking Pruning for Backdoor Mitigation: An Optimization Perspective
por: Li, Nan, et al.
Publicado: (2024)

Heterogeneous Graph Backdoor Attack
por: Chen, Jiawei, et al.
Publicado: (2025)

Backdoor Secrets Unveiled: Identifying Backdoor Data with Optimized Scaled Prediction Consistency
por: Pal, Soumyadeep, et al.
Publicado: (2024)

Trapping Attacker in Dilemma: Examining Internal Correlations and External Influences of Trigger for Defending GNN Backdoors
por: Yang, Fan, et al.
Publicado: (2026)

LoBAM: LoRA-Based Backdoor Attack on Model Merging
por: Yin, Ming, et al.
Publicado: (2024)

Your Agent Can Defend Itself against Backdoor Attacks
por: Changjiang, Li, et al.
Publicado: (2025)

Injecting Universal Jailbreak Backdoors into LLMs in Minutes
por: Chen, Zhuowei, et al.
Publicado: (2025)

How to Craft Backdoors with Unlabeled Data Alone?
por: Wang, Yifei, et al.
Publicado: (2024)

Pay Attention to the Triggers: Constructing Backdoors That Survive Distillation
por: De Muri, Giovanni, et al.
Publicado: (2025)

TooBadRL: Trigger Optimization to Boost Effectiveness of Backdoor Attacks on Deep Reinforcement Learning
por: Zhang, Mingxuan, et al.
Publicado: (2025)

GenBFA: An Evolutionary Optimization Approach to Bit-Flip Attacks on LLMs
por: Das, Sanjay, et al.
Publicado: (2024)

Data-Chain Backdoor: Do You Trust Diffusion Models as Generative Data Supplier?
por: Lu, Junchi, et al.
Publicado: (2025)

Backdoor Vectors: a Task Arithmetic View on Backdoor Attacks and Defenses
por: Pawlak, Stanisław, et al.
Publicado: (2025)

Winning the MIDST Challenge: New Membership Inference Attacks on Diffusion Models for Tabular Data Synthesis
por: Wu, Xiaoyu, et al.
Publicado: (2025)

Structure-Aware Distributed Backdoor Attacks in Federated Learning
por: Jian, Wang, et al.
Publicado: (2026)

Unlearn to Relearn Backdoors: Deferred Backdoor Functionality Attacks on Deep Learning Models
por: Shin, Jeongjin, et al.
Publicado: (2024)

UNIDOOR: A Universal Framework for Action-Level Backdoor Attacks in Deep Reinforcement Learning
por: Ma, Oubo, et al.
Publicado: (2025)

Backdoor Attack on Vertical Federated Graph Neural Network Learning
por: Yang, Jirui, et al.
Publicado: (2024)

Purifying Generative LLMs from Backdoors without Prior Knowledge or Clean Reference
por: Li, Jianwei, et al.
Publicado: (2026)

Stealthy Dual-Trigger Backdoors: Attacking Prompt Tuning in LM-Empowered Graph Foundation Models
por: Xue, Xiaoyu, et al.
Publicado: (2025)

ICLShield: Exploring and Mitigating In-Context Learning Backdoor Attacks
por: Ren, Zhiyao, et al.
Publicado: (2025)

BACKTIME: Backdoor Attacks on Multivariate Time Series Forecasting
por: Lin, Xiao, et al.
Publicado: (2024)

Invisible Backdoor Attack Through Singular Value Decomposition
por: Chen, Wenmin, et al.
Publicado: (2024)

Graph Neural Backdoor: Fundamentals, Methodologies, Applications, and Future Directions
por: Yang, Xiao, et al.
Publicado: (2024)

Towards Sample-specific Backdoor Attack with Clean Labels via Attribute Trigger
por: Zhu, Mingyan, et al.
Publicado: (2023)

Instruction Backdoor Attacks Against Customized LLMs
por: Zhang, Rui, et al.
Publicado: (2024)

Exploiting Layer-Specific Vulnerabilities to Backdoor Attack in Federated Learning
por: Foroughi, Mohammad Hadi, et al.
Publicado: (2026)

Backdoor Attacks on Fault Detection and Localization in Cyber-Physical Systems
por: Jean, Abile, et al.
Publicado: (2026)

Client-Side Patching against Backdoor Attacks in Federated Learning
por: Molina-Coronado, Borja
Publicado: (2024)

Guiding not Forcing: Enhancing the Transferability of Jailbreaking Attacks on LLMs via Removing Superfluous Constraints
por: Yang, Junxiao, et al.
Publicado: (2025)

PRUNE: A Patching Based Repair Framework for Certifiable Unlearning of Neural Networks
por: Li, Xuran, et al.
Publicado: (2025)

BadImplant: Injection-based Multi-Targeted Graph Backdoor Attack
por: Khan, Md Nabi Newaz, et al.
Publicado: (2026)

Backdoor Graph Condensation
por: Wu, Jiahao, et al.
Publicado: (2024)

Uncovering, Explaining, and Mitigating the Superficial Safety of Backdoor Defense
por: Min, Rui, et al.
Publicado: (2024)

POT: Inducing Overthinking in LLMs via Black-Box Iterative Optimization
por: Li, Xinyu, et al.
Publicado: (2025)