AI Safety

Can Textual Unlearning Solve Cross-Modality Safety Alignment? EMNLP 2024 Catastrophic Goodhart: regularizing RLHF with KL divergence does not mitigate heavy-tailed reward misspecification NeurIPS 2024 Catastrophic Jailbreak of Open-source LLMs via Exploiting Generation ICLR 2024 Certified Adversarial Robustness for Rate Encoded Spiking Neural Networks ICLR 2024 Certified Adversarial Robustness via Randomized α-Smoothing for Regression Models NeurIPS 2024 CGI-DM: Digital Copyright Authentication for Diffusion Models via Contrasting Gradient Inversion CVPR 2024 Chain-of-Verification Reduces Hallucination in Large Language Models ACL 2024 Challenging Large Language Models with New Tasks: A Study on their Adaptability and Robustness ACL 2024 Chameleon: Increasing Label-Only Membership Leakage with Adaptive Poisoning ICLR 2024 Chaos with Keywords: Exposing Large Language Models Sycophancy to Misleading Keywords and Evaluating Defense Strategies ACL 2024 Characterising Interventions in Causal Games UAI 2024 ChatGPT Doesn’t Trust Chargers Fans: Guardrail Sensitivity in Context EMNLP 2024 Chronic Poisoning: Backdoor Attack against Split Learning AAAI 2024 Circumventing Concept Erasure Methods For Text-To-Image Generative Models ICLR 2024 Citation: A Key to Building Responsible and Accountable Large Language Models NAACL 2024 CLAVE: An Adaptive Framework for Evaluating Values of LLM Generated Responses NeurIPS 2024 CleanGen: Mitigating Backdoor Attacks for Generation Tasks in Large Language Models EMNLP 2024 Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Models EACL 2024 CodeAttack: Revealing Safety Generalization Challenges of Large Language Models via Code Completion ACL 2024 Cognitive Bias in Decision-Making with LLMs EMNLP 2024 Cognitive Overload: Jailbreaking Large Language Models with Overloaded Logical Thinking NAACL 2024 COLD-Attack: Jailbreaking LLMs with Stealthiness and Controllability ICML 2024 COMBAT: Alternated Training for Effective Clean-Label Backdoor Attacks AAAI 2024 Composite Backdoor Attacks Against Large Language Models NAACL 2024 Compositional Curvature Bounds for Deep Neural Networks ICML 2024