LATIDIA · Investigación

MedConstBench: Evaluación comparativa de cómo las constituciones dirigen los LLM en ética médica

arXiv:2610.36513v1 Anuncio Tipo: nuevo Resumen: A medida que los grandes modelos de lenguaje avanzan rápidamente, se utilizan cada vez más para responder preguntas médicas, y muchos de estos modelos ahora se implementan localmente. Una vez que se implementa un modelo,

La noticia

arXiv:2610.36513v1 Announce Type: new Abstract: As large language models advance rapidly, they are increasingly used to answer medical questions, and many such models are now deployed locally. Once a model is deployed, its behavior is governed mainly by the ethical policy written into its constitution. However, most existing work focuses on factual correctness rather than the trade-off between competing ethical values. Even benchmarks that do evaluate ethics treat it as a fixed property of the model rather than as a variable, so how a constitution reshapes medical-ethics behavior remains largely unexplored. We introduce MedConstBench, a medical-ethics benchmark that treats the deployment constitution as a controlled variable, is organized around Beauchamp's