Cover Image

Understanding and Improving Continuous Adversarial Training for LLMs via In-context Learning Theory

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Fu, Shaopeng, Wang, Di
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Cryptography and Security
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Short-length Adversarial Training Helps LLMs Defend Long-length Jailbreak Attacks: Theoretical and Empirical Evidence
by: Fu, Shaopeng, et al.
Published: (2025)

Efficient Adversarial Training in LLMs with Continuous Attacks
by: Xhonneux, Sophie, et al.
Published: (2024)

Pre-trained Encoder Inference: Revealing Upstream Encoders In Downstream Machine Learning Services
by: Fu, Shaopeng, et al.
Published: (2024)

Improved Generation of Adversarial Examples Against Safety-aligned LLMs
by: Li, Qizhang, et al.
Published: (2024)

Understanding In-Context Learning of Linear Models in Transformers Through an Adversarial Lens
by: Anwar, Usman, et al.
Published: (2024)

Mitigating Error Amplification in Fast Adversarial Training
by: Zhao, Mengnan, et al.
Published: (2026)

Vulnerability-Aware Robust Multimodal Adversarial Training
by: Zhang, Junrui, et al.
Published: (2025)

Robustness Against Adversarial Attacks via Learning Confined Adversarial Polytopes
by: Hamidi, Shayan Mohajer, et al.
Published: (2024)

Improving Clean Accuracy via a Tangent-Space Perspective on Adversarial Training
by: Yi, Bongsoo, et al.
Published: (2024)

Robustness-Congruent Adversarial Training for Secure Machine Learning Model Updates
by: Angioni, Daniele, et al.
Published: (2024)

Generalization Properties of Adversarial Training for $\ell_0$-Bounded Adversarial Attacks
by: Delgosha, Payam, et al.
Published: (2024)

On the Effectiveness of Adversarial Training on Malware Classifiers
by: Bostani, Hamid, et al.
Published: (2024)

Understanding Deep Learning defenses Against Adversarial Examples Through Visualizations for Dynamic Risk Assessment
by: Echeberria-Barrio, Xabier, et al.
Published: (2024)

Adversarial Suffix Filtering: a Defense Pipeline for LLMs
by: Khachaturov, David, et al.
Published: (2025)

Adversarial Machine Unlearning
by: Di, Zonglin, et al.
Published: (2024)

FedBAP: Backdoor Defense via Benign Adversarial Perturbation in Federated Learning
by: Yan, Xinhai, et al.
Published: (2025)

On the Privacy Risk of In-context Learning
by: Duan, Haonan, et al.
Published: (2024)

Steering Beyond the Support: Adversarial Training on Unsupervised Jailbroken Activation Simulation
by: Chen, Luoyu, et al.
Published: (2026)

Passive Inference Attacks on Split Learning via Adversarial Regularization
by: Zhu, Xiaochen, et al.
Published: (2023)

Adversarial Attacks on Graph Neural Networks via Meta Learning
by: Zügner, Daniel, et al.
Published: (2019)

Understanding Sensitivity of Differential Attention through the Lens of Adversarial Robustness
by: Takahashi, Tsubasa, et al.
Published: (2025)

Detecting Adversarial Data via Provable Adversarial Noise Amplification
by: Mumcu, Furkan, et al.
Published: (2026)

Enhancing Adversarial Attacks via Parameter Adaptive Adversarial Attack
by: Jin, Zhibo, et al.
Published: (2024)

Learning Robust and Privacy-Preserving Representations via Information Theory
by: Zhang, Binghui, et al.
Published: (2024)

Elevating Defenses: Bridging Adversarial Training and Watermarking for Model Resilience
by: Thakkar, Janvi, et al.
Published: (2023)

CCLab: Adversarial Testing of Learning- and Non-Learning-Based Congestion Controllers
by: Chen, Zhi, et al.
Published: (2026)

Refusal-Trained LLMs Are Easily Jailbroken As Browser Agents
by: Kumar, Priyanshu, et al.
Published: (2024)

Using Graph Theory for Improving Machine Learning-based Detection of Cyber Attacks
by: Zonneveld, Giacomo, et al.
Published: (2024)

DeMem: Privacy-Enhanced Robust Adversarial Learning via De-Memorization
by: Luo, Xiaoyu, et al.
Published: (2024)

AdvSGM: Differentially Private Graph Learning via Adversarial Skip-gram Model
by: Zhang, Sen, et al.
Published: (2025)

Adaptive Meta-learning-based Adversarial Training for Robust Automatic Modulation Classification
by: Bamdad, Amirmohammad, et al.
Published: (2025)

Adversarial Prompt Evaluation: Systematic Benchmarking of Guardrails Against Prompt Input Attacks on LLMs
by: Zizzo, Giulio, et al.
Published: (2025)

Intrinsic Fingerprint of LLMs: Continue Training is NOT All You Need to Steal A Model!
by: Yoon, Do-hyeon, et al.
Published: (2025)

Trident: Improving Malware Detection with LLMs and Behavioral Features
by: Saul, Rebecca, et al.
Published: (2026)

Effective Universal Unrestricted Adversarial Attacks using a MOE Approach
by: Baia, A. E., et al.
Published: (2021)

MCP Safety Training: Learning to Refuse Falsely Benign MCP Exploits using Improved Preference Alignment
by: Halloran, John
Published: (2025)

Adversarial Contrastive Learning for LLM Quantization Attacks
by: Song, Dinghong, et al.
Published: (2026)

Temporal Analysis of Adversarial Attacks in Federated Learning
by: Mapakshi, Rohit, et al.
Published: (2025)

Introducing Adaptive Continuous Adversarial Training (ACAT) to Enhance ML Robustness
by: elShehaby, Mohamed, et al.
Published: (2024)

How to Enhance Downstream Adversarial Robustness (almost) without Touching the Pre-Trained Foundation Model?
by: Liu, Meiqi, et al.
Published: (2025)