Imagen de Portada

BadChain: Backdoor Chain-of-Thought Prompting for Large Language Models

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Xiang, Zhen, Jiang, Fengqing, Xiong, Zidi, Ramasubramanian, Bhaskar, Poovendran, Radha, Li, Bo
Formato:	Preprint
Publicado:	2024
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

CleanGen: Mitigating Backdoor Attacks for Generation Tasks in Large Language Models
por: Li, Yuetai, et al.
Publicado: (2024)

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

Game of Trojans: Adaptive Adversaries Against Output-based Trojaned-Model Detectors
por: Sahabandu, Dinuka, et al.
Publicado: (2024)

SoSBench: Benchmarking Safety Alignment on Six Scientific Domains
por: Jiang, Fengqing, et al.
Publicado: (2025)

ACE: A Model Poisoning Attack on Contribution Evaluation Methods in Federated Learning
por: Xu, Zhangchen, et al.
Publicado: (2024)

Double-Dip: Thwarting Label-Only Membership Inference Attacks with Transfer Learning and Randomization
por: Rajabi, Arezoo, et al.
Publicado: (2024)

DarkMind: Latent Chain-of-Thought Backdoor in Customized LLMs
por: Guo, Zhen, et al.
Publicado: (2025)

ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates
por: Jiang, Fengqing, et al.
Publicado: (2024)

Brave: Byzantine-Resilient and Privacy-Preserving Peer-to-Peer Federated Learning
por: Xu, Zhangchen, et al.
Publicado: (2024)

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

CANTXSec: A Deterministic Intrusion Detection and Prevention System for CAN Bus Monitoring ECU Activations
por: Donadel, Denis, et al.
Publicado: (2025)

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

BadScientist: Can a Research Agent Write Convincing but Unsound Papers that Fool LLM Reviewers?
por: Jiang, Fengqing, et al.
Publicado: (2025)

Backdooring Instruction-Tuned Large Language Models with Virtual Prompt Injection
por: Yan, Jun, et al.
Publicado: (2023)

Backdoored Retrievers for Prompt Injection Attacks on Retrieval Augmented Generation of Large Language Models
por: Clop, Cody, et al.
Publicado: (2024)

BadRSSD: Backdoor Attacks on Regularized Self-Supervised Diffusion Models
por: Wang, Jiayao, et al.
Publicado: (2026)

Under-confidence Backdoors Are Resilient and Stealthy Backdoors
por: Peng, Minlong, et al.
Publicado: (2022)

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

Fusing Pruned and Backdoored Models: Optimal Transport-based Data-free Backdoor Mitigation
por: Lin, Weilin, 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)

BadFU: Backdoor Federated Learning through Adversarial Machine Unlearning
por: Lu, Bingguang, et al.
Publicado: (2025)

BadThink: Triggered Overthinking Attacks on Chain-of-Thought Reasoning in Large Language Models
por: Liu, Shuaitong, et al.
Publicado: (2025)

RigorLLM: Resilient Guardrails for Large Language Models against Undesired Content
por: Yuan, Zhuowen, et al.
Publicado: (2024)

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

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

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

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

Unreal Thinking: Chain-of-Thought Hijacking via Two-stage Backdoor
por: Chang, Wenhan, et al.
Publicado: (2026)

Model Supply Chain Poisoning: Backdooring Pre-trained Models via Embedding Indistinguishability
por: Wang, Hao, et al.
Publicado: (2024)

Chain-of-Thought Prompting of Large Language Models for Discovering and Fixing Software Vulnerabilities
por: Nong, Yu, et al.
Publicado: (2024)

SequentialBreak: Large Language Models Can be Fooled by Embedding Jailbreak Prompts into Sequential Prompt Chains
por: Saiem, Bijoy Ahmed, et al.
Publicado: (2024)

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

Prompt Obfuscation for Large Language Models
por: Pape, David, et al.
Publicado: (2024)

Backdooring Masked Diffusion Language Models
por: Cao, Daniel Yiming, et al.
Publicado: (2026)

Self-Purification Mitigates Backdoors in Multimodal Diffusion Language Models
por: Wan, Guangnian, et al.
Publicado: (2026)

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

The 'Sure' Trap: Multi-Scale Poisoning Analysis of Stealthy Compliance-Only Backdoors in Fine-Tuned Large Language Models
por: Tan, Yuting, et al.
Publicado: (2025)

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

Conscious Data Contribution via Community-Driven Chain-of-Thought Distillation
por: Libon, Lena, et al.
Publicado: (2025)

Krait: A Backdoor Attack Against Graph Prompt Tuning
por: Song, Ying, et al.
Publicado: (2024)