Peer Reviewed

Music plays an important role in the personal fulfillment and cognitive performance of deaf and hard of hearing (DHH) individuals. Since deafness is a spectrum -- as are DHH individuals' preferences and perceptions of music -- a more situated understanding of their interaction with music is needed. To understand the music experience of this population, we conducted social media analyses, both qualitatively and quantitatively, in the deaf and hard of hearing Reddit communities and followed this up with interviews with DHH individuals. Our analysis revealed accessibility challenges, i.e., hearing aids were not customized for music, visual/haptic techniques developed were not widely available and affordable, and accessibility accommodations by music streaming apps and offline events were imperfect -- leading to suboptimal music experiences. In response, DHH individuals leveraged audio, visual, and physical senses to listen to the music -- treating it as a full-body experience; as accessibility heuristics, they may also prefer familiar, non lyrical, instrument-heavy, or loud music, which was perceived as more accessible. The DHH community embodied mutual support among music lovers, evidenced by active information sharing around music. Misconceptions by hearing individuals regarding how DHH individuals listen to music were reported, which is a major hurdle for creating a more accessible music experience. We reflect on design justice for DHH individuals' music experience based on a situated understanding and propose practical design implications to create a more accessible music experience for them.

Prompting LLMs for complex tasks (e.g., building a trip advisor chatbot) needs humans to clearly articulate customized requirements (e.g., “start the response with a tl;dr”). However, existing prompt engineering instructions often lack focused training on requirement articulation and instead tend to emphasize increasingly automatable strategies (e.g., tricks like adding role-plays and “think step-by-step”). To address the gap, we introduce Requirement-Oriented Prompt Engineering ( ROPE ), a paradigm that focuses human attention on generating clear, complete requirements during prompting. We implement ROPE through an assessment and training suite that provides deliberate practice with LLM-generated feedback. In a randomized controlled experiment with 30 novices, ROPE significantly outperforms conventional prompt engineering training (20% vs. 1% gains), a gap that automatic prompt optimization cannot close. Furthermore, we demonstrate a direct correlation between the quality of input requirements and LLM outputs. Our work paves the way to empower more end-users to build complex LLM applications.