safety alignment

ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates AAAI 2025 Confusion is the Final Barrier: Rethinking Jailbreak Evaluation and Investigating the Real Misuse Threat of LLMs EMNLP 2025 Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements ICLR 2025 COSMIC: Generalized Refusal Direction Identification in LLM Activations ACL 2025 COVER: Context-Driven Over-Refusal Verification in LLMs ACL 2025 Data to Defense: The Role of Curation in Aligning Large Language Models Against Safety Compromise EMNLP 2025 Do as I do (Safely): Mitigating Task-Specific Fine-tuning Risks in Large Language Models ICLR 2025 Don’t Say No: Jailbreaking LLM by Suppressing Refusal ACL 2025 Do We Really Need Curated Malicious Data for Safety Alignment in Multi-modal Large Language Models? CVPR 2025 DREAM: Disentangling Risks to Enhance Safety Alignment in Multimodal Large Language Models NAACL 2025 Durable Quantization Conditioned Misalignment Attack on Large Language Models ICLR 2025 Efficient Safety Alignment of Large Language Models via Preference Re-ranking and Representation-based Reward Modeling ACL 2025 Emerging Safety Attack and Defense in Federated Instruction Tuning of Large Language Models ICLR 2025 English as Defense Proxy: Mitigating Multilingual Jailbreak via Eliciting English Safety Knowledge EMNLP 2025 Exploring Intrinsic Alignments Within Text Corpus AAAI 2025 FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts AAAI 2025 Fine-Tuning Lowers Safety and Disrupts Evaluation Consistency ACL 2025 Forewarned is Forearmed: Pre-Synthesizing Jailbreak-like Instructions to Enhance LLM Safety Guardrail to Potential Attacks EMNLP 2025 How to Fine-Tune Safely on a Budget: Model Adaptation Using Minimal Resources EMNLP 2025 Immune: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment CVPR 2025 Improving LLM Safety Alignment with Dual-Objective Optimization ICML 2025 Injecting Universal Jailbreak Backdoors into LLMs in Minutes ICLR 2025 Jailbreak Attack Initializations as Extractors of Compliance Directions EMNLP 2025 Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks ICLR 2025 Jailbreak-Tuning: Models Efficiently Learn Jailbreak Susceptibility EMNLP 2025