Researchers have developed a new method called perturbation profiles to better understand how AI models interpret scanpath data. This technique involves systematically altering specific properties of scanpaths, such as fixation order or duration, and observing the impact on classification accuracy. The study profiled nine different scanpath descriptors across multiple datasets, revealing sensitivities and robust features, with a Python package released to facilitate this analysis. AI
IMPACT Provides a novel method for evaluating the interpretability of AI models in analyzing human gaze data.
RANK_REASON The cluster contains an academic paper detailing a new methodology for AI analysis. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
