AI Safety

Shielding in Resource-Constrained Goal POMDPs AAAI 2023 Single Image Backdoor Inversion via Robust Smoothed Classifiers CVPR 2023 SlowLiDAR: Increasing the Latency of LiDAR-Based Detection Using Adversarial Examples CVPR 2023 Sparse Black-Box Multimodal Attack for Vision-Language Adversary Generation EMNLP 2023 Speak, Memory: An Archaeology of Books Known to ChatGPT/GPT-4 EMNLP 2023 Spoq: Scaling Machine-Checkable Systems Verification in Coq OSDI 2023 SSDA: Secure Source-Free Domain Adaptation ICCV 2023 State-wise Safe Reinforcement Learning: A Survey IJCAI 2023 Static and Sequential Malicious Attacks in the Context of Selective Forgetting NeurIPS 2023 Stay Moral and Explore: Learn to Behave Morally in Text-based Games ICLR 2023 SteerLM: Attribute Conditioned SFT as an (User-Steerable) Alternative to RLHF EMNLP 2023 Step by Step Loss Goes Very Far: Multi-Step Quantization for Adversarial Text Attacks EACL 2023 StereoMap: Quantifying the Awareness of Human-like Stereotypes in Large Language Models EMNLP 2023 Stop Uploading Test Data in Plain Text: Practical Strategies for Mitigating Data Contamination by Evaluation Benchmarks EMNLP 2023 Structure Invariant Transformation for better Adversarial Transferability ICCV 2023 Survival Instinct in Offline Reinforcement Learning NeurIPS 2023 Systematic Rectification of Language Models via Dead-end Analysis ICLR 2023 Tackling Safe and Efficient Multi-Agent Reinforcement Learning via Dynamic Shielding (Student Abstract) AAAI 2023 TAILCHECK: A Lightweight Heap Overflow Detection Mechanism with Page Protection and Tagged Pointers OSDI 2023 Targeted Data Generation: Finding and Fixing Model Weaknesses ACL 2023 Targeted Knowledge Infusion To Make Conversational AI Explainable and Safe AAAI 2023 The Best Defense Is a Good Offense: Adversarial Augmentation Against Adversarial Attacks CVPR 2023 The Dark Side of Dynamic Routing Neural Networks: Towards Efficiency Backdoor Injection CVPR 2023 The #DNN-Verification Problem: Counting Unsafe Inputs for Deep Neural Networks IJCAI 2023 The Past, Present and Better Future of Feedback Learning in Large Language Models for Subjective Human Preferences and Values EMNLP 2023