AI Safety

Tougher Text, Smarter Models: Raising the Bar for Adversarial Defence Benchmarks COLING 2025 Towards Adversarial Robustness via Debiased High-Confidence Logit Alignment ICCV 2025 Towards an Automated Framework to Audit Youth Safety on TikTok EMNLP 2025 Towards a Theory of AI Personhood AAAI 2025 Towards Better Value Principles for Large Language Model Alignment: A Systematic Evaluation and Enhancement ACL 2025 Towards Certification of Uncertainty Calibration under Adversarial Attacks ICLR 2025 Towards Computational Foreseeability AAAI 2025 Towards Effective and Sparse Adversarial Attack on Spiking Neural Networks via Breaking Invisible Surrogate Gradients CVPR 2025 Towards Enforcing Company Policy Adherence in Agentic Workflows EMNLP 2025 Towards Fully Exploiting LLM Internal States to Enhance Knowledge Boundary Perception ACL 2025 Towards Holistic Evaluation of Large Audio-Language Models: A Comprehensive Survey EMNLP 2025 Towards Reverse Engineering of Language Models: A Survey EMNLP 2025 Towards Reward Fairness in RLHF: From a Resource Allocation Perspective ACL 2025 Towards Robust Knowledge Unlearning: An Adversarial Framework for Assessing and Improving Unlearning Robustness in Large Language Models AAAI 2025 Towards Safety Reasoning in LLMs: AI-agentic Deliberation for Policy-embedded CoT Data Creation ACL 2025 Towards Statistical Factuality Guarantee for Large Vision-Language Models EMNLP 2025 Towards Truly Open, Language-Specific, Safe, Factual, and Specialized Large Language Models COLING 2025 Towards Trustworthy Federated Learning with Untrusted Participants ICML 2025 Towards Trustworthy Machine Learning Under Distribution Shifts AAAI 2025 Towards Trustworthy Summarization of Cardiovascular Articles: A Factuality-and-Uncertainty-Aware Biomedical LLM Approach EMNLP 2025 Towards Understanding the Fragility of Multilingual LLMs against Fine-Tuning Attacks NAACL 2025 Training Data Provenance Verification: Did Your Model Use Synthetic Data from My Generative Model for Training? CVPR 2025 Training Verification-Friendly Neural Networks via Neuron Behavior Consistency AAAI 2025 Transferable Direct Prompt Injection via Activation-Guided MCMC Sampling EMNLP 2025 Transforming Healthcare Decision Making Using Artificial Intelligence AAAI 2025