AI人格分裂实锤，30万道送命题，撕开OpenAI、谷歌「遮羞布」
"道德推理" 相关文章
MoReBench: Evaluating Procedural and Pluralistic Moral Reasoning in Language Models, More than Outcomes
Deliberative Dynamics and Value Alignment in LLM Debates
Deliberative Dynamics and Value Alignment in LLM Debates
One Model, Many Morals: Uncovering Cross-Linguistic Misalignments in Computational Moral Reasoning
Ethics-Based Refusals Without Ethics-Based Refusal Training
MORABLES: A Benchmark for Assessing Abstract Moral Reasoning in LLMs with Fables
One more reason for AI capable of independent moral reasoning: alignment itself and cause prioritisation
Emergent morality in AI weakens the Orthogonality Thesis
Beyond Ethical Alignment: Evaluating LLMs as Artificial Moral Assistants
Normative Moral Pluralism for AI: A Framework for Deliberation in Complex Moral Contexts
"Pull or Not to Pull?'': Investigating Moral Biases in Leading Large Language Models Across Ethical Dilemmas
"Just a strange pic": Evaluating 'safety' in GenAI Image safety annotation tasks from diverse annotators' perspectives
MoralBench: Moral Evaluation of LLMs
Token and Taboo
Token and Taboo
GPT-4o竟是「道德专家」？解答50道难题，比纽约大学教授更受欢迎
研究发现 OpenAI 的 GPT-4o 道德推理能力胜过人类专家
