AI Safety

Pancreatic Cancer Diagnosis System AAAI 2025 PANDAS: Improving Many-shot Jailbreaking via Positive Affirmation, Negative Demonstration, and Adaptive Sampling ICML 2025 PatchGuard: Adversarially Robust Anomaly Detection and Localization through Vision Transformers and Pseudo Anomalies CVPR 2025 Path Drift in Large Reasoning Models: How First-Person Commitments Override Safety EMNLP 2025 PBI-Attack: Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for Toxicity Maximization EMNLP 2025 PBI-Attack: Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for Toxicity Maximization NAACL 2025 PD3F: A Pluggable and Dynamic DoS-Defense Framework against resource consumption attacks targeting Large Language Models EMNLP 2025 PEARL: Towards Permutation-Resilient LLMs ICLR 2025 People will agree what I think: Investigating LLM’s False Consensus Effect NAACL 2025 Permitted Knowledge Boundary: Evaluating the Knowledge-Constrained Responsiveness of Large Language Models EMNLP 2025 Persistent Pre-training Poisoning of LLMs ICLR 2025 Persona-judge: Personalized Alignment of Large Language Models via Token-level Self-judgment ACL 2025 Persuasion Dynamics in LLMs: Investigating Robustness and Adaptability in Knowledge and Safety with DuET-PD EMNLP 2025 Persuasion for Social Good: How to Build and Break AI AAAI 2025 Phi: Preference Hijacking in Multi-modal Large Language Models at Inference Time EMNLP 2025 Pierce the Mists, Greet the Sky: Decipher Knowledge Overshadowing via Knowledge Circuit Analysis EMNLP 2025 PIGDreamer: Privileged Information Guided World Models for Safe Partially Observable Reinforcement Learning ICML 2025 PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization ACL 2025 PIGuard: Prompt Injection Guardrail via Mitigating Overdefense for Free ACL 2025 Pixel2Feature Attack (P2FA): Rethinking the Perturbed Space to Enhance Adversarial Transferability ICML 2025 PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preference ACL 2025 PLA: Prompt Learning Attack against Text-to-Image Generative Models ICCV 2025 Playing the Fool: Jailbreaking LLMs and Multimodal LLMs with Out-of-Distribution Strategy CVPR 2025 PL-Guard: Benchmarking Language Model Safety for Polish ACL 2025 Pluralistic Alignment for Healthcare: A Role-Driven Framework EMNLP 2025