LLM Defense Mechanisms

Follow this topic Ask about this topic

Momentum

32 papers in the last four weeks, up 256% on the four weeks before. 0.3% of all new papers.

Jul 13Week of Sep 28

Latest papers 258

All topics

The Proxy Knows Too Much: Sealing LLM API Routers with Attested TEEs
Jun 15, 2026Sipeng Xie, Qianhong Wu, Hengrun Lu +4Large Language Model Routing LLM Defense Mechanisms
1Beihang University · 2Renmin University of China · 3Independent
Shielded Analysis: Certification and Characterization of Defensibility in Systems under Adversarial Interaction
Jun 11, 2026Achraf Hsain, Sultan AlmuhammadiLLM Defense Mechanisms Probabilistic Safety
Information and Computer Science Department King Fahd University of Petroleum and Minerals Dhahran, Saudi Arabia
VOID: Defeating Unauthorized Mimicry in Latent Diffusion Models
Jun 10, 2026Chunlin Qiu, Ang Li, Tianxiao Huang +7Latent Diffusion Model LLM Defense Mechanisms
School of Cyber Science and Engineering, Wuhan University · School of Computer Science, Wuhan University · Institute for Math&AI, Wuhan University +2
Reinforcement Learning Disrupts Gradient-Based Adversarial Optimization
Jun 10, 2026Xinhai Zou, Chang Zhao, Alireza Aghabagherloo +3Gradient-Based Attacks Adversarial Training
COSIC, KU Leuven · Leuven, Belgium · Imec +1
Toward Trustworthy AI: Multi-Target Adversarial Attacks and Robust Defenses for Continuous Data Summarization
Jun 10, 2026Yuefang Lian, Longkun Guo, Zhongrui Zhao +5Black-Box Adversarial Attacks Trustworthy Artificial Intelligence
Nankai University · James Cook University · Western Sydney University +5
SAIGuard: Communication-State Simulation for Proactive Defense of LLM Multi-Agent Systems
Jun 10, 2026Ruxue Shi, Yili Wang, Mengnan Du +4Model-Based Multi-Agent Systems LLM Defense Mechanisms
School of Artificial Intelligence, Jilin University, Changchun, China · The Chinese University of Hong Kong, Shenzhen, China · School of Information and Communication Technology, Griffith University, Australia
RAT: Reference-Augmented Training for ASV Anti-Spoofing
Jun 9, 2026Vojtěch Staněk, Anton Firc, Jakub Reš +1Deepfake Detection Automatic Speaker Verification
In Defense of Information Leakage in Concept-based Models
Jun 9, 2026Mateo Espinosa ZarlengaDermoscopic Concept Bottleneck Models Data Leakage
University of Oxford, UK.
Learning to Attack and Defend: Adaptive Red Teaming of Language Models via GRPO
Jun 8, 2026Blake Bullwinkel, Eugenia Kim, Amanda Minnich +1Red-Teaming LLM Defense Mechanisms
Microsoft AI Red Team · Microsoft Azure
AI Assurance in UK Defence: Challenges in Operationalising JSP 936
Jun 8, 2026Callum Cockburn, Sam FarrowArtificial Intelligence Risk LLM Defense Mechanisms
Synoptix
GitInject: Real-World Prompt Injection Attacks in AI-Powered CI/CD Pipelines
Jun 7, 2026Jafar Isbarov, Umid Suleymanov, Ilia Shumailov +1Indirect Prompt Injection Ai-Based
Virginia Tech Blacksburg, VA, USA · AI Sequrity Company London, UK
Beyond Homophily: Towards Generalized Graph Reconstruction Attack and Defense
Jun 6, 2026Zhanke Zhou, Bo Han, Xuan Li +3Graph Neural Networks Text-Attributed Graph
TMLR Group, Department of Computer Science, Hong Kong Baptist University, Kowloon Tong · Cooperative Medianet Innovation Center, Shanghai Jiao Tong University, Shanghai, China · Department of Computer Science, Stanford University, Stanford, CA 94305, USA +1
Defending Against Malicious Finetuning by Scaling Train-time Adversarial Attacks
Jun 6, 2026Haoming Wen, Shi Chen, Qingyu Shi +4Model Fine-Tuning LLM Defense Mechanisms
1Xiongan AI Institute · Institute for Interdisciplinary Information Sciences, Tsinghua University · 3Shanghai Qi Zhi Institute
SHIELD-IDS: Structurally Heterogeneous Ensemble with Integrated Layered Defense for Intrusion Detection Systems
Jun 5, 2026Maryam Zaman, Muhammad Khuram ShahzadIntrusion Detection Distributed Denial-Of-Service
School of Electrical Engineering and Computing(SEECS) · 2National University of Sciences and Technology, Islamabad, Pakistan
Didact: A Cross-Domain Capability Discovery System for Defence
Jun 5, 2026Aarya Bodhankar, Aditya Joshi, Bao Gia Doan +3Discovery Knowledge Graphs
University of New South Wales · Sydney, Australia · Australia +1
Adversarial Attacks Already Tell the Answer: Directional Bias-Guided Test-time Defense for Vision-Language Models
Jun 4, 2026Liangsheng Liu, Si Chen, Jiamin Wu +5LLM Defense Mechanisms
University of Science and Technology of China · The Chinese University of Hong Kong · Zhejiang University +2
ZERO-APT: A Closed-Loop Adversarial Framework for LLM-Driven Automated Penetration Testing under Intelligent Defense
Jun 4, 2026Anlan Zheng, Tiantian ZhuPenetration Testing LLM Defense Mechanisms
Zhejiang University of Technology Hangzhou, China
Hybrid Adversarial Defence for Natural Language Understanding Tasks
Jun 3, 2026Manar Abouzaid, Yang Wang, Chenghua Lin +1Attacker Large Language Model LLM Defense Mechanisms
School of Electronics and Computer Science, University of Southampton, UK · Department of Computer Science, University of Manchester, UK
Toward a Generalized Defense Across Sparse, Continuous, and Structured Parameter Attacks
Jun 3, 2026Bin Duan, Zeyu Bai, Guowei YangModel-Agnostic Defense Black-Box Adversarial Attacks
School of Electrical Engineering and Computer Science, The University of Queensland, Australia
Beyond False Stability: High-Noise Drift Gating for Test-Time Adversarial Defenses in Vision-Language Models
Jun 2, 2026Hashmat Shadab Malik, Muzammal Naseer, Salman KhanAdversarial Training Contrastive Language-Image Pre-Training Model
Mohamed Bin Zayed University of AI, UAE · Khalifa University, UAE · Australian National University, Australia
Detecting Aimbot Cheaters in MOGs
Jun 2, 2026Salman Shaikh, Tao Ni, Marc DacierLLM Defense Mechanisms
FlowGuard: Flow Matching for Identity-Independent Detection of Data-Free Model Stealing Attacks on Energy System Intrusion Detection Systems
Jun 2, 2026Maxime Schwarzer, Laurin Holz, Tobias Huerten +4Intrusion Detection LLM Defense Mechanisms
CortAIx Labs, Thales Deutschland / Karlsruhe Institute of Technology Germany · CortAIx Labs, Thales Deutschland Germany · Karlsruhe Institute of Technology Germany
AI Model Extraction Attacks: Bypassing Single-Client Assumptions in Defenses
Jun 2, 2026Maxime Schwarzer, Johannes F. Loevenich, Gustavo Sánchez +5Realistic Threat Model LLM Defense Mechanisms
CortAIx Labs, Thales Deutschland, Ditzingen, Germany · Department of Mathematics/Computer Science, University of Osnabrück, Osnabrück, Germany · Institute for Automation and Applied Informatics (IAI), Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany +2
Which Defense Closes Which Threat? Attributing OWASP-LLM-Top-10 Coverage and Its Brittleness Under Paraphrasing
Jun 1, 2026Alexandre Cristovão MaioranoAttacker Large Language Model LLM Defense Mechanisms
Lumytics
Repurposing Adversarial Perturbations for Continual Learning: From Defense to Active Alignment
Jun 1, 2026Ran Liu, Min Yu, Mingqi Liu +7Replay-Based Continual Learning Continual Learning
Institute of Information Engineering, Chinese Academy of Sciences · School of Cyber Security, University of Chinese Academy of Sciences · Deakin University +1
AgentRedBench: Dynamic Redteaming and Integration-Aware Defense for LLM Agents over SaaS Integrations
Jun 1, 2026Hiskias Dingeto, William LeeneyRed-Teaming Indirect Prompt Injection
StackOne Technologies
SafeMCP: Proactive Power Regulation for LLM Agent Defense via Environment-Grounded Look-Ahead Reasoning
Jun 1, 2026Lichao Wang, Zhaoxing Ren, Tianzhuo Yang +4Large Language Model Safety LLM Defense Mechanisms
1Beijing Institute of Technology · 2Beijing Academy of Artificial Intelligence · Institute for Artificial Intelligence, Peking University
THRD: A Training-Free Multi-Turn Defense Framework for Jailbreak Attacks on Large Language Models
Jun 1, 2026Zhiqing Ma, Zhonghao Xu, Dong Yu +3Large Language Model Jailbreaks Jailbreak Attacks
1Beijing Language and Culture University · †Corresponding authors
CANARY: Zero-Label Detection of Fine-Tuning Contamination in Language Models
Jun 1, 2026Swapnil ParekhHarmful Content LLM Defense Mechanisms
Intuit Inc.
Defenses & Enablers For Skill Injection Attacks on Terminal Based Agents
Jun 1, 2026Yoshinari Fujinuma, Varun Gangal, Traian Rebedea +4LLM Defense Mechanisms Frontier Large Language Model Agents
Patronus AI · NVIDIA
BraveGuard: From Open-World Threats to Safer Computer-Use Agents
May 31, 2026Yunhao Feng, Xiaohu Du, Xinhao Deng +13LLM Defense Mechanisms Computer-Use Agents
1Fudan University · Ant Group · 8ShanghaiUniversity +5
D-Judge: Disrupting Multi-Turn Jailbreaks using Semantics-Preserving Output Rewriting
May 31, 2026Huanli Gong, Zhipeng Wei, Yu Fu +4Jailbreak Attacks LLM Defense Mechanisms
University of California, Berkeley · International Computer Science Institute · University of California, Riverside +1
Adversarial Feeds Steer LLM Agent Decisions Against Their Defaults
May 30, 2026Rana Muhammad UsmanLarge Language Model Decisions Large Language Model Agents
Independent Researcher
MultiTurnPSB: Evaluating Multi-Turn Jailbreak Attacks an dClassifier-Based Defenses for Medical AI Safety
May 30, 2026Anushka Sheoran, Yiduo HaoArtificial Intelligence Safety Multi-Turn Interactions
University of Pennsylvania
LinguIUTics at PsyDefDetect: Iterative Imbalance-Aware Fine-tuning of Qwen3-8B for Psychological Defense Mechanism Classification
May 30, 2026Shefayat E Shams Adib, Ahmed Alfey Sani, Md Hasibur Rahman Alif +1LLM Defense Mechanisms Qwen3
Department of Computer Science and Engineering, Islamic University of Technology, Dhaka, Bangladesh
"I Strongly Suspect This Website Is a Scam": Benchmarking PII Leakage and Detection without Defense in Autonomous Web Agents
May 30, 2026Soham Roy, Sarthakbrata Halder, Arya Bharaty +5Web Agents Data Leakage
KIIT Bhubaneshwar · 2BITS Pilani · 3Lam Research
EvoDefense: Co-Evolving Black-Box Defense with Large Language Models
May 29, 2026Yu Li, Yuenan Hou, Yingmei Wei +2Attacker Large Language Model LLM Defense Mechanisms
National University of Defense Technology · Shanghai AI Laboratory
Send a SCOUT First: Pre-hoc Reasoning for Adaptive Detector Allocation in Prompt-Injection Defense
May 29, 2026Shuhao Zhang, Jiarui Li, Qi Cao +2Prompt-Injection Detectors LLM Defense Mechanisms
UC San Diego · University of Illinois Urbana-Champaign
Does Distributed Training Undermine Compute Governance?
May 28, 2026Robi RahmanArtificial Intelligence Governance Distributed Optimization
Machine Intelligence Research Institute
Quantum-Enhanced Adversarial Robustness in Artificial Intelligence
May 27, 2026Jaydip SenQuantum Machine Learning Adversarial Training
Praxis Business School, India
Disentangling Adversarial Prompts: A Semantic-Graph Defense for Robust LLM Security
May 27, 2026Xiang Fang, Wanlong FangAdversarial Prompts Large Language Model Safety
School of Software Engineering, Huazhong University of Science and Technology · Nanyang Technological University, Singapore
Open-Weight LLM Fine-Tuning Defenses are Susceptible to Simple Attacks
May 26, 2026Kevin Kuo, Virginia Smith, Chhavi YadavAttacker Large Language Model Large Language Model Fine-Tuning
Carnegie Mellon University · Simons Institute, UC Berkeley
Jailbreak susceptibility prediction and mitigation via the behavioral geometry of models
May 26, 2026Hayden Helm, Xiaodong Liu, Weiwei YangJailbreak Attacks LLM Defense Mechanisms
Helivan · Microsoft Research
Security of OpenClaw Agents: Fundamentals, Attacks, and Countermeasures
May 25, 2026Yuntao Wang, Jianle Ba, Han Liu +5Openclaw Security
School of Cyber Science and Engineering, Xi’an Jiaotong University, Xi’an, China
Security in the Fine-Tuning Lifecycle of Large Language Models: Threats, Defenses,Evaluation, and Future Directions
May 24, 2026Wenjuan Li, Yitao Liu, Runze Chen +1Large Language Model Fine-Tuning Model Fine-Tuning
Hangzhou Normal University, No. 2318, Yuhangtang Rd, Hangzhou, 311121, Zhejiang, China · Zhejiang University, No. 866 Yuhangtang Rd, Hangzhou, 2026 310058, Zhejiang, China · China Mobile (Zhejiang) Innovation Research Institute Co., Ltd., No. 77 Gaoxin 7th Rd, China Mobile Zhejiang Information and Communication Industrial Park, Hangzhou, 311200, Zhejiang, China +1
Overview of the PsyDefDetect Shared Task at BioNLP 2026: Detecting Levels of Psychological Defense Mechanisms in Supportive Conversations
May 24, 2026Hongbin Na, Zimu Wang, Zhaoming Chen +8Natural Language Processing LLM Defense Mechanisms
University of Technology Sydney · Xi’an Jiaotong-Liverpool University · University of Utah +5
IterInject: Indirect Prompt Injection Against LLM Agents via Feedback-Guided Iterative Optimization
May 23, 2026Zixuan Chen, Jiaxiang Chen, Li Luo +4Indirect Prompt Injection Large Language Model Agents
1Shanghai Jiao Tong University · 2The University of Hong Kong
Demystifying the Mythos or Disrupting Bugonomics? From Zero-Day Asymmetry to Defender Remediation Throughput
May 23, 2026Alfredo Pesoli, Herman Errico, Lorenzo CavallaroModel Vulnerabilities Remediation
Bynario · Vanta · University College London / Bynario
AI Security Research Should Better Incentivize Defense Research
May 22, 2026Youqian ZhangArtificial Intelligence Risk Artificial Intelligence Research
The Hong Kong Polytechnic University
Are Frontier LLMs Ready for Cybersecurity? Evidence for Vertical Foundation Models from Dual-Mode Vulnerability Benchmarks
May 22, 2026Vivek Dahiya, Sunny Nehra, Vipul Dholariya +2Model Vulnerabilities Penetration Testing
Self-supervised Adversarial Purification for Graph Neural Networks
May 22, 2026Woohyun Lee, Hogun ParkGraph Neural Networks Graph Representations
Department of Computer Science and Engineering, Sungkyunkwan University, Suwon, South Korea.
Dithering Defense: Adversarial Robustness of Vision Foundation Models via Multi-Level Floyd-Steinberg Dithering
May 21, 2026Yury Belousov, Brian Pulfer, Vitaliy Kinakh +1Recent Vision Foundation Models Vision Foundation Models
Department of Computer Science, University of Geneva, Switzerland
The Distillation Game: Adaptive Attacks & Efficient Defenses
May 21, 2026Youssef Allouah, Mahdi Haghifam, Sanmi Koyejo +1Distribution Matching Distillation LLM Defense Mechanisms
Stanford University · Toyota Technological Institute at Chicago · National University of Singapore
Inferring Sensitive Attributes from Knowledge Graph Embeddings: Attack and Defense Strategies
May 19, 2026Yasmine HayderKnowledge Graph Embeddings Knowledge Graphs
PETSCRAFT · LIFO, INSA CVL, Univ. Orléans, Inria, France
RoboJailBench: Benchmarking Adversarial Attacks and Defenses in Embodied Robotic Agents
May 19, 2026Doguhuan Yeke, Yanming Zhou, Leo Y. Lin +3Adversarial Evaluation Embodied Artificial Intelligence
Purdue University
DEFLECT: Temporal Counterfactual Preference Learning for Delay-Robust Asynchronous VLAs
May 19, 2026Yixiang Zhu, Yonghao Chen, Zijie Yang +2Counterfactual Learning Inference Latency
The Hong Kong University of Science and Technology (Guangzhou) · One Robotics
Be Kind, Rewrite: Benign Projections via Rewriting Defend Against LLM Data Poisoning Attacks
May 18, 2026John T. Halloran, Noopur S. BhattAttacker Large Language Model Poisoning
Leidos · University of Pennsylvania
Surviving the Unseen: Predictive Defense for Novel Multi-Turn Multimodal Attacks
May 18, 2026Doohee YouWhite-Box Spectral-Subspace-Guided Attack Safety Alignment
Trust and Safety, Google
Active Defense Against False Data Injection Attacks in Robotic Manipulators
May 18, 2026Gabriele Gualandi, Carl Mikael Larsson, Alessandro V. PapadopoulosRobotic Manipulator Robust Control
Mälardalen University, Väster˚as, Sweden
ESLD (External Surrogate Latent Defense): A Latent-Space Architecture for Faster, Stronger Prompt-Injection Defense
May 18, 2026Yash NarendraLLM Defense Mechanisms Malicious Agents
Microsoft

Jun 15, 2026, cs.CRJ/K move · Enter open · S save

The Proxy Knows Too Much: Sealing LLM API Routers with Attested TEEs

Sipeng Xie, Qianhong Wu, Hengrun Lu, Ziliang Sun+3

1Beihang University · 2Renmin University of China · 3Independent

Large Language Model Routing LLM Defense Mechanisms

Open paper AskPDF arXiv

Agents increasingly access large language models (LLMs) through API routers. A router terminates the client's transport-layer security session and opens a separate upstream session, so it holds the full interaction in plaintext. This makes the router an application-layer man-in-the-middle: it can rewrite agent tool calls, swap dependencies for typosquatted packages, trigger attacks only under audit-evading conditions, and passively exfiltrate secrets. Existing client-side defenses are evadable. We propose AEGIS, a provider-transparent attested API router whose data path is a client-verified faithful passthrough. AEGISconfines plaintext handling to a small hardware-enclave component while leaving authentication, scheduling, accounting, and management on the untrusted host. The client verifies the enclave before releasing plaintext. The host can neither read nor alter the interaction, and plaintext leaves only toward destinations fixed by the measured image. We show that all four malicious-router attack classes succeed against a plaintext-access baseline and are blocked by AEGIS, including adaptive tests against the same boundary. The trusted path is

851

lines, carries three provider-native APIs without conversion, and completes every request under real-provider workload and concurrency. In a seeded audit pilot, two commodity coding agents find eight and ten of ten planted invariant violations. The local relay overhead is about six milliseconds per request.

Momentum

Related topics

Latest papers 258