AI Safety

Stable Unlearnable Example: Enhancing the Robustness of Unlearnable Examples via Stable Error-Minimizing Noise AAAI 2024 STAR: SocioTechnical Approach to Red Teaming Language Models EMNLP 2024 State-wise safe reinforcement learning with pixel observations L4DC 2024 Stealth edits to large language models NeurIPS 2024 Stealthy and Persistent Unalignment on Large Language Models via Backdoor Injections NAACL 2024 Stealthy Attack on Large Language Model based Recommendation ACL 2024 Steering Llama 2 via Contrastive Activation Addition ACL 2024 Steganographic Passport: An Owner and User Verifiable Credential for Deep Model IP Protection Without Retraining CVPR 2024 Stepwise Alignment for Constrained Language Model Policy Optimization NeurIPS 2024 STOP! Benchmarking Large Language Models with Sensitivity Testing on Offensive Progressions EMNLP 2024 Stress-Testing Capability Elicitation With Password-Locked Models NeurIPS 2024 Stronger and Transferable Node Injection Attacks AAAI 2024 Subspace Defense: Discarding Adversarial Perturbations by Learning a Subspace for Clean Signals COLING 2024 Subtle Biases Need Subtler Measures: Dual Metrics for Evaluating Representative and Affinity Bias in Large Language Models ACL 2024 Subtle Signatures, Strong Shields: Advancing Robust and Imperceptible Watermarking in Large Language Models ACL 2024 Successfully Guiding Humans with Imperfect Instructions by Highlighting Potential Errors and Suggesting Corrections EMNLP 2024 SuperDeepFool: a new fast and accurate minimal adversarial attack NeurIPS 2024 Supporting Upper Elementary Students in Learning AI Concepts with Story-Driven Game-Based Learning AAAI 2024 Survival of the Safest: Towards Secure Prompt Optimization through Interleaved Multi-Objective Evolution EMNLP 2024 Taming False Positives in Out-of-Distribution Detection with Human Feedback AISTATS 2024 Task-Agnostic Detector for Insertion-Based Backdoor Attacks NAACL 2024 Taxonomy Driven Fast Adversarial Training AAAI 2024 Teaching Language Models to Self-Improve by Learning from Language Feedback ACL 2024 Teach LLMs to Phish: Stealing Private Information from Language Models ICLR 2024 Temporal-Distributed Backdoor Attack against Video Based Action Recognition AAAI 2024