Backdoor Defense

Follow this topic Ask about this topic

Momentum

4 papers in the last four weeks, up 33% on the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 37

All topics

Backdoor Containment via Expert Quarantine and Shutdown in LLMs
Sep 30, 2026Jianwei Li, Min-Seon Kim, Jung-Eun KimLLM Backdoor Attacks Backdoor Defense in LLMs
Department of Computer Science North Carolina State University Raleigh, NC 27606, USA
Selective Channel Restoration for Backdoored Vision-Language Models
Sep 29, 2026Shuming Liu, Zhifang Zhang, Suqin Yuan +3VLM Robustness Large Vision-Language Models
Southeast University · The University of Queensland · University of Sydney +1
Backdoor as Probe: Test-Time Adversarial Defense for CLIP
Sep 28, 2026Zhongqi Wang, Jie Zhang, Nie Sen +3VLM Robustness Backdoor Defense
Key Laboratory of AI Safety of CAS, Institute of Computing Technology, Chinese Academy of Sciences (CAS), Beijing, China · University of Chinese Academy of Sciences, Beijing, China · Xuzhou University of Technology, China
ODPure: Backdoor Purification for Object Detection via Ensemble Corruption Consensus
Sep 23, 2026Li Zeng, Mingcheng Duan, Longfei Fan +5Backdoor Attacks Backdoor Defense
School of Computer and Communication Engineering, Changsha University of Science & Technology, Changsha 410114, China · School of Computer Science, Xiangtan University, Xiangtan 411105, China · School of Software, Yunnan University, Kunming 650500, China +3
Beyond Small Patches: Black-Box Detection and Purification of Diverse Backdoor Triggers
Sep 2, 2026Ahmed Abdelnaby, Mohamed ElmahallawyBackdoor Detection Backdoor Defense
School of Engineering and Applied Sciences, Washington State University, Richland, WA 99354, USA
Backdoor Decontamination Dynamics in LLM Agents
Aug 11, 2026Gabriel Huang, Abhay Puri, Léo Boisvert +4LLM Backdoor Attacks Backdoor Defense in LLMs
ServiceNow Research · Mila – Quebec AI Institute · Polytechnique Montréal +3
DiffSafeMerge: Mitigating Backdoor Inheritance in Diffusion Model Merging
Aug 10, 2026Jiayang Zhang, Ji Guo, Jiachen Li +2Backdoor Attacks Backdoor Defense
University of Electronic Science and Technology of China, Chengdu, China · Wuhan University of Technology, Wuhan, China
Mitigating Backdoors via Decoy Shortcuts and Knowledge Decoupling
Aug 1, 2026Zixuan Zhu, Rui Wang, Lihua Jing +1Backdoor Attacks Backdoor Defense
Institute of Information Engineering, Chinese Academy of Sciences · School of Cyber Security, University of Chinese Academy of Sciences
Anti-Backdoor Coreset Selection via Cumulative Entropy
Jul 28, 2026Qi Zhao, Christian WressneggerNeural Network Robustness Training Data Selection
KASTEL Security Research Labs, Karlsruhe Institute of Technology (KIT)
SpeechGuard: Online Defense against Backdoor Attacks on Speech Recognition Models
Jul 17, 2026Jinwen Xin, Xixiang LvBackdoor Detection Automatic Speech Recognition
School of Cyber Engineering Xidian University Xi’an, China
ORAN-DEFEND: Subspace Detection and Sanitization of Backdoor DRL xApps in Open RAN
Jul 7, 2026Md Raihan Uddin, Fatemeh Lotfi, Tolunay Seyfi +1Adversarial Attacks Backdoor Detection
Department of Electrical and Computer Engineering, Clemson University, Clemson, SC, USA
The Power of Backdoor Absorption in Community Training
Jul 7, 2026Issam Seddik, Sami Souihi, Mohamed Tamaazousti +1Algorithmic Auditing Backdoor Attacks
Université Paris-Saclay, CEA LIST, Palaiseau, France
Two Sides of the Same Coin: Learning the Backdoor to Remove the Backdoor
Jul 7, 2026Qi Zhao, Christian WressneggerBackdoor Attacks Backdoor Detection
KASTEL Security Research Labs · Karlsruhe Institute of Technology (KIT)
Wireless Backdoor Attack and Defense for Semantic Communications over Multiple Access Channel
Jun 29, 2026Yalin E. Sagduyu, Tugba Erpek, Aylin Yener +1Backdoor Attacks Semantic Communication
Nexcepta, Gaithersburg, MD, USA · The Ohio State University, Columbus, OH, USA · University of Maryland, College Park, MD, USA
Clustering Unsupervised Representations as Defense against Poisoning Attacks on Speech Commands Classification System
Jun 27, 2026Thomas Thebaud, Sonal Joshi, Henry Li +4Backdoor Attacks Automatic Speech Recognition
Center for Language and Speech Processing, Johns Hopkins University, USA · Faculty of Information Technology, Brno University of Technology, Czechia
Rethinking Molecular Graph Backdoors under Chemistry-aware Admission
Jun 22, 2026Thinh T. H. Nguyen, Sze Jue Yang, Khoa D. Doan +2Graph Neural Networks Backdoor Attacks
VinUniversity, Hanoi, Vietnam · Universiti Malaya
InstantForget: Update-Free Backdoor Unlearning with Inference-Time Feature Reset
Jun 14, 2026Zhenyu YuBackdoor Defense Machine Unlearning
College of Computer Science and Artificial Intelligence Fudan University
Rethinking Backdoor Adversarial Unlearning through the Lens of Catastrophic Forgetting in Continual Learning
Jun 12, 2026Zhenqian Zhu, Yamin Hu, Yujiang Liu +5Backdoor Defense Machine Unlearning
Harbin Institute of Technology, Shenzhen Shenzhen, Guangdong, China · Shenzhen Key Laboratory of Media Security, Shenzhen University Shenzhen, Guangdong, China
PolicyGuard: Towards Test-time and Step-level Adversary (Backdoor) Defense for Reinforcement Learning Agent
Jun 11, 2026Junfeng Guo Heng HuangBackdoor Attacks Backdoor Defense
Computer Science, University of Maryland College Park, USA.
Dummy Backdoor as a Defense: Removing Unknown Backdoors via Shared Internal Mechanisms for Generative LLMs
Jun 10, 2026Kazuki Iwahana, Masaru Matsubayashi, Takuma Koyama +3LLM Backdoor Attacks Backdoor Defense in LLMs
1NTT Social Informatics Laboratories · 2Tohoku University
Backdoor Unlearning Generalization: A Path Toward the Removal of Unknown Triggers in LLMs
Jun 2, 2026Lisa Bouger, Théo Lasnier, Philippe Loubet Moundi +2LLM Backdoor Attacks Backdoor Defense in LLMs
1Inria Paris · 2Sorbonne Université · 3Thales CDI
Patcher: Post-Hoc Patching of Backdoored Large Language Models
Jun 2, 2026Anjun Gao, Yueyang Quan, Yufei Xia +2LLM Backdoor Attacks Backdoor Attacks
University of Louisville · University of North Texas
From Prompt Injection to Persistent Control: Defending Agentic Harness Against Trojan Backdoors
May 29, 2026Jiejun Tan, Zhicheng Dou, Xinyu Yang +4Prompt Injection LLM Agent Security
Gaoling School of Artificial Intelligence, Renmin University of China
Token-Level Generalization in LoRA Adapter Backdoors: Attack Characterization and Behavioral Detection
May 28, 2026Travis LelleLLM Backdoor Attacks Backdoor Detection
TimeGuard: Channel-wise Pool Training for Backdoor Defense in Time Series Forecasting
May 21, 2026Quang Duc Nguyen, Siyuan Liang, Yiming Li +2Backdoor Attacks Time Series Forecasting
College of Computing and Data Science, Nanyang Technological University, Singapore.
Provable Robustness against Backdoor Attacks via the Primal-Dual Perspective on Differential Privacy
May 20, 2026Aman Saxena, Jan Schuchardt, Yan Scholten +1Randomized Smoothing Adversarial Robustness
Department of Computer Science, Technical University of Munich · Munich Data Science Institute · MCML +1
Trusted Weights, Treacherous Optimizations? Optimization-Triggered Backdoor Attacks on LLMs
May 20, 2026Yifei Wang, Yida Yang, Tianlin Li +3LLM Backdoor Attacks Backdoor Attacks
Shanghai Jiao Tong University · Tongji University · Beihang University +1
Your Neighbors Know: Leveraging Local Neighborhoods for Backdoor Detection in Decentralized Learning
May 19, 2026Sayan Biswas, Antoine Boutet, Davide Frey +7Backdoor Detection Decentralized Learning
1EPFL · 2Inria, INSA Lyon, CITI · 3Univ. Rennes, Inria, CNRS, IRISA
Be Kind, Rewrite: Benign Projections via Rewriting Defend Against LLM Data Poisoning Attacks
May 18, 2026John T. Halloran, Noopur S. BhattLLM Backdoor Attacks Training Data Curation
Leidos · University of Pennsylvania
Lightweight and Fast Backdoor Model Detection
May 17, 2026Yinbo Yu, Jing Fang, Xuewen Zhang +4Backdoor Attacks Backdoor Detection
College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing, Jiangsu, 210016, China · Shenzhen Research Institute of Northwestern Polytechnical University, Shenzhen, Guangdong, 518057, China · School of Software, Northwestern Polytechnical University, Xi’an, Shaanxi, 710072, P.R. China +2

Sep 30, 2026, cs.AIJ/K move · Enter open · S save

Backdoor Containment via Expert Quarantine and Shutdown in LLMs

Jianwei Li, Min-Seon Kim, Jung-Eun Kim

Department of Computer Science North Carolina State University Raleigh, NC 27606, USA

LLM Backdoor Attacks Backdoor Defense in LLMs Backdoor Defense

Open paper AskPDF arXiv

Backdoored large language models (LLMs) can behave normally on benign inputs while producing attacker-specified outputs under hidden triggers. Existing defenses span four stages--prior-training, in-training, post-training, and inference-time--and share one of two underlying strategies: either suppress backdoor learning (by filtering poisoned data or interrupting its acquisition during optimization) or learn, then purify (by repairing model weights or gating inputs after a fully backdoored model has formed). We propose a third strategy, learn, but channel: allow backdoor formation during training but route it into a designated, quarantined component that can be disabled at deployment. To this end, we propose Quarantined Expert Shutdown QES, a computationally efficient containment strategy built in a regularization-steered MoE-like setting. Specifically, given a poisoned dataset, QES augments a Transformer-based language model with routed expert-specific LoRA branches and lightweight routers, and uses auxiliary routing objectives to attract trigger-conditioned behavior into a designated expert while preserving benign capability elsewhere. At deployment, mitigation reduces to a single constant-time operation: zeroing the quarantined expert's routing weight, without trigger screening or further updating model weights. Empirically, our methods reduce the attack success rate ASR from 100% to 0-10% on most settings across two tasks, three attacks, and four model families, while downstream utility is often preserved or only modestly affected. These results establish learn, but channel as a previously unexplored regime for backdoor containment in generative LLMs.

Momentum

Related topics

Latest papers 37