AI Safety

CHASE: Contextual History for Adaptive and Simple Exploitation in Large Language Model Jailbreaking AAAI 2026 Clean-Label Physical Backdoor Attacks with Data Distillation AAAI 2026 Closed-Form Concept Erasure via Double Projections CVPR 2026 CoCoNUTS: Concentrating on Content while Neglecting Uninformative Textual Styles for AI-Generated Peer Review Detection ACL 2026 Co-Designing Unplugged Learning Activities with K-2 Teachers for Early AI Literacy Education AAAI 2026 Code-Switching as a Safety Failure Mode in Large Language Models: An Empirical Study of Roman Urdu across English, Mixed, and Transliteration-Only Inputs EACL 2026 Composable Assurance for AI Alignment: A Framework for Propagating Formal Safety Properties Through MLOps AAAI 2026 ConfGuard: A Simple and Effective Backdoor Detection for Large Language Models AAAI 2026 Confirmation Bias: A Challenge for Scalable Oversight AAAI 2026 Conformal Feedback Alignment: Quantifying Answer-Level Reliability for Robust LLM Alignment EACL 2026 Consensus Learning with Multi-Party Perturbation Triggers for Secure Model Access AAAI 2026 Control Illusion: The Failure of Instruction Hierarchies in Large Language Models AAAI 2026 ConVerse: Benchmarking Contextual Safety in Agent-to-Agent Conversations EACL 2026 Copyright Detective: A Forensic System to Evidence LLMs Flickering Copyright Leakage Risks ACL 2026 CoSPED: Consistent Soft Prompt Targeted Data Extraction and Defense AAAI 2026 Cost-Minimized Label-Flipping Poisoning Attack to LLM Alignment AAAI 2026 CRADLE Bench: A Clinician-Annotated Benchmark for Multi-Faceted Mental Health Crisis and Safety Risk Detection EACL 2026 Creating Blank Canvas Against AI-enabled Image Forgery AAAI 2026 DAVSP: Safety Alignment for Large Vision-Language Models via Deep Aligned Visual Safety Prompt AAAI 2026 Debiased Dual-Invariant Defense for Adversarially Robust Person Re-Identification AAAI 2026 Delphi: A Neuro-Symbolic Framework for Individualized, Safe and Interpretable Treatment Recommendation AAAI 2026 Designing Incident Reporting Systems for Harms from General-Purpose AI AAAI 2026 Detecting Compute Structuring in AI Governance Is Likely Feasible AAAI 2026 Detection of Adversarial Prompts with Model Predictive Entropy EACL 2026 Diagnosing and Repairing Unsafe Channels in Vision-Language Models via Causal Discovery and Dual-Modal Safety Subspace Projection CVPR 2026