Can LLMs create lasting flashcards from readers' highlights?
The report tests whether LLMs can convert readers' highlights into durable, spaced-repetition flashcards. It evaluates model accuracy, conciseness, and learning utility, finding usable but variable results across models and highlight complexity.