Research Design

by Sohail Khan. Language models are becoming integral to daily life as trusted sources of advice. While their utility has expanded from simple tasks like text summarization to more complex interactions, the empathetic quality of their responses is crucial. This article explores methods to assess the emotional appropriateness of these models, using metrics such as BLEU, ROUGE, and Sentence Transformers. By analyzing models like LLaMA in mental health dialogues, we learn that while they suffer through traditional word-based metrics, LLaMA's performance in capturing empathy through semantic similarity is promising. In addition, we must advocate for continuous monitoring to ensure these models support their users' mental well-being effectively.

by Mango Jane Angar. As we celebrate Disability Awareness Month at the D-Lab alongside the UC Berkeley scholarly community, how can we, as social scientists, individually promote accessibility and inclusion? To advance accessibility, we should focus on addressing the barriers faced by individuals with disabilities, using our research to provide insights for effective policy recommendations. Although most of us do not focus on disability-related issues, including disability as a demographic characteristic in our data collection can greatly enhance our understanding of diverse populations and improve the comprehensiveness of our analyses. This small step can contribute to broader efforts toward inclusion and social equity.