Imagen de Portada

DarkMind: Latent Chain-of-Thought Backdoor in Customized LLMs

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Guo, Zhen, Shi, Shanghao, Yazdani, Shamim, Zhang, Ning, Tourani, Reza
Formato:	Preprint
Publicado:	2025
Materias:	Cryptography and Security Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

TraceGuard: Process-Guided Firewall against Reasoning Backdoors in Large Language Models
por: Guo, Zhen, et al.
Publicado: (2026)

Persistent Backdoor Attacks in Continual Learning
por: Guo, Zhen, et al.
Publicado: (2024)

BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models
por: Xiang, Zhen, et al.
Publicado: (2024)

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

Unveiling the Unseen: Exploring Whitebox Membership Inference through the Lens of Explainability
por: Li, Chenxi, et al.
Publicado: (2024)

BoBa: Boosting Backdoor Detection through Data Distribution Inference in Federated Learning
por: Jiang, Zhengyuan, et al.
Publicado: (2024)

Low Rank Adaptation for Adversarial Perturbation
por: Liu, Han, et al.
Publicado: (2026)

A Generative Framework for Low-Cost Result Validation of Machine Learning-as-a-Service Inference
por: Kumar, Abhinav, et al.
Publicado: (2023)

Trusting What You Cannot See: Auditable Fine-Tuning and Inference for Proprietary AI
por: Jin, Heng, et al.
Publicado: (2026)

ProFLingo: A Fingerprinting-based Intellectual Property Protection Scheme for Large Language Models
por: Jin, Heng, et al.
Publicado: (2024)

Authority Backdoor: A Certifiable Backdoor Mechanism for Authoring DNNs
por: Yang, Han, et al.
Publicado: (2025)

Thought-Transfer: Indirect Targeted Poisoning Attacks on Chain-of-Thought Reasoning Models
por: Chaudhari, Harsh, et al.
Publicado: (2026)

Robust Backdoor Removal by Reconstructing Trigger-Activated Changes in Latent Representation
por: Iwahana, Kazuki, et al.
Publicado: (2025)

Backdoor Channels Hidden in Latent Space: Cryptographic Undetectability in Modern Neural Networks
por: Eggen, Marte, et al.
Publicado: (2026)

Echoes within the Reasoning: Stealthy and Effective Watermarking via Chain of Thought
por: Lu, Jiacheng, et al.
Publicado: (2026)

Rethinking Jailbreak Detection of Large Vision Language Models with Representational Contrastive Scoring
por: Hua, Peichun, et al.
Publicado: (2025)

Turn-Based Structural Triggers: Prompt-Free Backdoors in Multi-Turn LLMs
por: Lu, Yiyang, et al.
Publicado: (2026)

Unlearning Backdoor Attacks through Gradient-Based Model Pruning
por: Dunnett, Kealan, et al.
Publicado: (2024)

CBD: A Certified Backdoor Detector Based on Local Dominant Probability
por: Xiang, Zhen, et al.
Publicado: (2023)

Countering Backdoor Attacks in Image Recognition: A Survey and Evaluation of Mitigation Strategies
por: Dunnett, Kealan, et al.
Publicado: (2024)

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

LATTEO: A Framework to Support Learning Asynchronously Tempered with Trusted Execution and Obfuscation
por: Kumar, Abhinav, et al.
Publicado: (2025)

BackdoorBench: A Comprehensive Benchmark and Analysis of Backdoor Learning
por: Wu, Baoyuan, et al.
Publicado: (2024)

DiffusionHijack: Supply-Chain PRNG Backdoor Attack on Diffusion Models and Quantum Random Number Defense
por: You, Ziyang, et al.
Publicado: (2026)

Inference-Time Backdoors via Chat Templates: From LLM Supply Chains to Agentic System Compromise
por: Fogel, Ariel, et al.
Publicado: (2026)

Multi-Trigger Poisoning Amplifies Backdoor Vulnerabilities in LLMs
por: Sivapiromrat, Sanhanat, et al.
Publicado: (2025)

Uncovering and Understanding FPR Manipulation Attack in Industrial IoT Networks
por: Ahsan, Mohammad Shamim, et al.
Publicado: (2026)

Backdoor Learning Curves: Explaining Backdoor Poisoning Beyond Influence Functions
por: Cinà, Antonio Emanuele, et al.
Publicado: (2021)

DeepSweep: An Evaluation Framework for Mitigating DNN Backdoor Attacks using Data Augmentation
por: Qiu, Han, et al.
Publicado: (2020)

Robustness Inspired Graph Backdoor Defense
por: Zhang, Zhiwei, et al.
Publicado: (2024)

On the Out-of-Distribution Backdoor Attack for Federated Learning
por: Xu, Jiahao, et al.
Publicado: (2025)

Trusted Weights, Treacherous Optimizations? Optimization-Triggered Backdoor Attacks on LLMs
por: Wang, Yifei, et al.
Publicado: (2026)

Hardware-Triggered Backdoors
por: Möller, Jonas, et al.
Publicado: (2026)

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

BadMerging: Backdoor Attacks Against Model Merging
por: Zhang, Jinghuai, et al.
Publicado: (2024)

Output Supervision Can Obfuscate the Chain of Thought
por: Drori, Jacob, et al.
Publicado: (2025)

Fusing Pruned and Backdoored Models: Optimal Transport-based Data-free Backdoor Mitigation
por: Lin, Weilin, et al.
Publicado: (2024)

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

PECAN: A Deterministic Certified Defense Against Backdoor Attacks
por: Zhang, Yuhao, et al.
Publicado: (2023)

Rethinking Graph Backdoor Attacks: A Distribution-Preserving Perspective
por: Zhang, Zhiwei, et al.
Publicado: (2024)