AI Safety

PrivacyRestore: Privacy-Preserving Inference in Large Language Models via Privacy Removal and Restoration ACL 2025 ProAdvPrompter: A Two-Stage Journey to Effective Adversarial Prompting for LLMs ICLR 2025 Probabilistic Aggregation and Targeted Embedding Optimization for Collective Moral Reasoning in Large Language Models ACL 2025 Probabilistic Shielding for Safe Reinforcement Learning AAAI 2025 Probe before You Talk: Towards Black-box Defense against Backdoor Unalignment for Large Language Models ICLR 2025 ProcessBench: Identifying Process Errors in Mathematical Reasoning ACL 2025 Profiling LLM’s Copyright Infringement Risks under Adversarial Persuasive Prompting EMNLP 2025 Programming Refusal with Conditional Activation Steering ICLR 2025 Progressively Label Enhancement for Large Language Model Alignment ICML 2025 Prompt-Character Divergence: A Responsibility Compass for Human-AI Creative Collaboration NeurIPS 2025 Prompt-Guided Internal States for Hallucination Detection of Large Language Models ACL 2025 PromptKeeper: Safeguarding System Prompts for LLMs EMNLP 2025 PROOD: A Simple LLM Out-of-Distribution Guardrail Leveraging Response Semantics EMNLP 2025 PROSAC: Provably Safe Certification for Machine Learning Models under Adversarial Attacks AAAI 2025 ProSec: Fortifying Code LLMs with Proactive Security Alignment ICML 2025 Protecting against simultaneous data poisoning attacks ICLR 2025 Protecting Model Adaptation from Trojans in the Unlabeled Data AAAI 2025 PROTECT: Policy-Related Organizational Value Taxonomy for Ethical Compliance and Trust ACL 2025 Prototype Guided Backdoor Defense via Activation Space Manipulation ICCV 2025 Provably Robust Explainable Graph Neural Networks against Graph Perturbation Attacks ICLR 2025 Proxy Barrier: A Hidden Repeater Layer Defense Against System Prompt Leakage and Jailbreaking EMNLP 2025 PTTA: Purifying Malicious Samples for Test-Time Model Adaptation ICML 2025 QA‐LIGN: Aligning LLMs through Constitutionally Decomposed QA EMNLP 2025 QGuard:Question-based Zero-shot Guard for Multi-modal LLM Safety ACL 2025 Qorǵau: Evaluating Safety in Kazakh-Russian Bilingual Contexts ACL 2025