Researchers have developed a video-based protocol to elicit explainable AI (XAI) requirements from stroke survivors, addressing the challenge of communicating complex AI concepts to individuals with communication impairments. The study involved three stroke survivors, two with aphasia, and three caregivers, utilizing scaffolding techniques like analogical bridging and projective personas to gather diverse XAI needs. The research also identified potential facilitation biases, such as normative bias and hypothesis confirmation bias, offering guidelines to mitigate these risks in future applications. AI
IMPACT This research could lead to more trustworthy human-machine systems by improving how AI explainability needs are gathered from vulnerable user groups.
RANK_REASON The item is a research paper detailing a new methodology for eliciting AI requirements. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Daily Papers →
- aphasia
- Caregivers and Veterans Omnibus Health Services Act of 2010
- Hugging Face
- human-machine systems
- patient
- Stroke Survivors and Caregivers Using an Online Mindfulness-based Intervention Together
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
