AI Safety

Does Physical Adversarial Example Really Matter to Autonomous Driving? Towards System-Level Effect of Adversarial Object Evasion Attack ICCV 2023 Don’t Retrain, Just Rewrite: Countering Adversarial Perturbations by Rewriting Text ACL 2023 Do Stochastic Parrots have Feelings Too? Improving Neural Detection of Synthetic Text via Emotion Recognition EMNLP 2023 Dynamic Generative Targeted Attacks With Pattern Injection CVPR 2023 Effective Ambiguity Attack Against Passport-Based DNN Intellectual Property Protection Schemes Through Fully Connected Layer Substitution CVPR 2023 Efficient Certified Training and Robustness Verification of Neural ODEs ICLR 2023 Efficient Global Robustness Certification of Neural Networks via Interleaving Twin-Network Encoding (Extended Abstract) IJCAI 2023 Efficient Verification of Neural Networks Against LVM-Based Specifications CVPR 2023 Eliminating Adversarial Noise via Information Discard and Robust Representation Restoration ICML 2023 Enforcing Hard Constraints with Soft Barriers: Safe Reinforcement Learning in Unknown Stochastic Environments ICML 2023 Enhancing Adversarial Robustness via Score-Based Optimization NeurIPS 2023 Enhancing Generalization of Universal Adversarial Perturbation through Gradient Aggregation ICCV 2023 Escaping saddle points in zeroth-order optimization: the power of two-point estimators ICML 2023 Ethical By Designer - How to Grow Ethical Designers of Artificial Intelligence (Extended Abstract) IJCAI 2023 Ethical Reasoning over Moral Alignment: A Case and Framework for In-Context Ethical Policies in LLMs EMNLP 2023 Evaluating Model-Free Reinforcement Learning toward Safety-Critical Tasks AAAI 2023 Explanation-based Finetuning Makes Models More Robust to Spurious Cues ACL 2023 Explanation-Guided Reward Alignment IJCAI 2023 Exploiting Connections between Lipschitz Structures for Certifiably Robust Deep Equilibrium Models NeurIPS 2023 Exploring Practical Vulnerabilities of Machine Learning-based Wireless Systems NSDI 2023 Exploring the Relationship Between Architectural Design and Adversarially Robust Generalization CVPR 2023 Extremely Simple Activation Shaping for Out-of-Distribution Detection ICLR 2023 Failing with Grace: Learning Neural Network Controllers that are Boundedly Unsafe L4DC 2023 Failure-Resistant Intelligent Interaction for Reliable Human-AI Collaboration AAAI 2023 Fake the Real: Backdoor Attack on Deep Speech Classification via Voice Conversion Interspeech 2023