Researchers have developed and validated an eye-tracking dataset to study how second-language learners process idiomatic expressions. The dataset captures cognitive costs associated with literal-first processing in L2 English speakers of varying proficiency levels, using data from Portuguese native speakers. Preliminary analysis shows a strong link between language proficiency and regressive eye movements, suggesting the dataset can serve as a benchmark for evaluating human processing models and large language models' figurative understanding. AI
IMPACT Provides a new benchmark for evaluating the figurative understanding capabilities of large language models.
RANK_REASON This is a research paper presenting a new dataset for evaluating language processing models.
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
