AI Safety

Feature Separation and Recalibration for Adversarial Robustness CVPR 2023 FedGame: A Game-Theoretic Defense against Backdoor Attacks in Federated Learning NeurIPS 2023 Few-shot Backdoor Attacks via Neural Tangent Kernels ICLR 2023 FLIP: A Provable Defense Framework for Backdoor Mitigation in Federated Learning ICLR 2023 Fooling SHAP with Stealthily Biased Sampling ICLR 2023 FORK: A Bite-Sized Test Set for Probing Culinary Cultural Biases in Commonsense Reasoning Models ACL 2023 Foveate, Attribute, and Rationalize: Towards Physically Safe and Trustworthy AI ACL 2023 From Robustness to Privacy and Back ICML 2023 Fundamental Limits in Formal Verification of Message-Passing Neural Networks ICLR 2023 Gradient-Informed Neural Network Statistical Robustness Estimation AISTATS 2023 Guiding LLM to Fool Itself: Automatically Manipulating Machine Reading Comprehension Shortcut Triggers EMNLP 2023 Guiding Safe Exploration with Weakest Preconditions ICLR 2023 Hallucination Detection for Generative Large Language Models by Bayesian Sequential Estimation EMNLP 2023 Holistic Adversarial Robustness of Deep Learning Models AAAI 2023 Honesty Is the Best Policy: Defining and Mitigating AI Deception NeurIPS 2023 How to Choose your Best Allies for a Transferable Attack? ICCV 2023 Human Control: Definitions and Algorithms UAI 2023 Identification of the Adversary from a Single Adversarial Example ICML 2023 Identifying and Adapting Transformer-Components Responsible for Gender Bias in an English Language Model EMNLP 2023 Ignore This Title and HackAPrompt: Exposing Systemic Vulnerabilities of LLMs Through a Global Prompt Hacking Competition EMNLP 2023 Improving Adversarial Robustness of Masked Autoencoders via Test-time Frequency-domain Prompting ICCV 2023 Improving Adversarial Robustness via Information Bottleneck Distillation NeurIPS 2023 Improving Robustness with Adaptive Weight Decay NeurIPS 2023 Improving the robustness of NLI models with minimax training ACL 2023 Improving the Transferability of Adversarial Samples by Path-Augmented Method CVPR 2023