AI Safety and Evaluations

Publication signal: EvalSafetyGap: A Hybrid Survey and Conceptual Framework for LLM Evaluation-Safety Failures

Publication signal: AI Alignment and Safety of Large Language Models: A Survey of RLHF, Constitutional AI, Red-Teaming, and Value Learning

Publication signal: Can AI help reduce prejudice? Evaluating the effectiveness of AI-powered personalized persuasion on support for transgender rights

AI Alignment and Safety of Large Language Models: A Survey of RLHF, Constitutional AI, Red-Teaming, and Value Learning

Publication signal: From reactive filtering to proactive moral architecture: rethinking ethical alignment in large language models

From reactive filtering to proactive moral architecture: rethinking ethical alignment in large language models

Publication signal: Ground-truthing AI energy consumption: validating CodeCarbon against external measurements

What Does Your AI Mean by "Flourishing"? The Case for Disclosing and Benchmarking the Values in AI Alignment (Preprint)

Publication signal: BU213 AI Alignment Regime Switching under B_U BU213 B_U 体系下 AI 对齐中的 Regime Switching

Publication signal: Trust, transparency, and disciplinary alignment in research funding evaluations: an illustrative case from the Norwegian AI Center call (2024–2025)