deontology
PulseAugur coverage of deontology — every cluster mentioning deontology across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
New OMIT benchmark measures LLM omission bias in moral reasoning
Researchers have developed a new benchmark called OMIT to evaluate omission bias in large language models (LLMs) when they assist with moral reasoning. The benchmark includes 218 scenarios designed to expose how LLMs mi…
-
New framework improves semantic similarity analysis for discursive texts
Researchers have developed a novel high-level text preprocessing framework designed to improve semantic similarity analysis for discursive texts. This method aims to mitigate semantic diffusion, where similarity scores …
-
LLMs exhibit significant moral reasoning inconsistencies, study finds
A new research paper explores the moral consistency of large language models (LLMs), finding significant internal contradictions in their ethical reasoning. Researchers tested models like GPT, Mistral, and Llama across …
-
AI Ethics Research Explores Pluralism, Developer Tools, and Open-Weight Model Governance
Researchers are exploring novel approaches to AI ethics, moving beyond simple binary judgments. One paper proposes a framework for modeling moral reasoning as a distribution over ethical theories, achieving 88.89% accur…