Search Papers | Poster Sessions | All Posters
Poster B46 in Poster Session B - Thursday, August 8, 2024, 1:30 – 3:30 pm, Johnson Ice Rink
Common Sense Reasoning about Source Credibility
Peiyao Hu1 (), Mark K. Ho1; 1Stevens Institute of Technology
We often rely on others’ testimony to learn about new topics, such as health benefits of a novel food. However, the sources are not always knowledgeable, helpful, or unbiased, necessitating an assessment of their credibility. Here, we present a Bayesian model of source credibility, where a listener reasons about the expertise and intentions of a source. We consider a scenario where both parties have noisy access to the ground truth of familiar topics, which serves as a basis for rational inference about a source’s credibility on novel topics. This approach provides a computational framework for understanding how people respond to information in domains like science communication and media consumption.
Keywords: Social Learning Social Epistemology Theory of Mind Probabilistic Programming