Low-Cost AI Method Generates Reading Questions in Multiple Languages
Researchers have developed a cheaper, faster way to create reading comprehension questions that works across languages without requiring massive training datasets. The method could help schools and edtech companies provide personalized learning at scale, especially in languages where AI training data is scarce.
Originaltitel: Quinductor: A multilingual data-driven method for generating reading-comprehension questions using Universal Dependencies
<p>We propose a multilingual data-driven method for generating reading comprehension questions using dependency trees. Our method provides a strong, deterministic and inexpensive-to-train baseline for less-resourced languages. While a language-specific corpus is still required, its size is nowhere near those required by modern neural question generation (QG) architectures. Our method surpasses QG baselines previously reported in the literature in terms of automatic evaluation metrics and shows a good performance in terms of human evaluation.</p>