# Beyond English-Only Reading Comprehension: Experiments in Zero-Shot Multilingual Transfer for Bulgarian

arXiv: https://arxiv.org/abs/1908.01519 · HTML: https://mlsanity.com/paper/1908.01519/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 Aug 2019.

## Built on by (most cited first, 10 of 10)

- [The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants](https://mlsanity.com/paper/2308.16884/) (2023)
- [English Machine Reading Comprehension Datasets: A Survey](https://mlsanity.com/paper/2101.10421/) (2021)
- [Distractor Generation in Multiple-Choice Tasks: A Survey of Methods, Datasets, and Evaluation](https://mlsanity.com/paper/2402.01512/) (2024)
- [VNHSGE: VietNamese High School Graduation Examination Dataset for Large Language Models](https://mlsanity.com/paper/2305.12199/) (2023)
- [Leaf: Multiple-Choice Question Generation](https://mlsanity.com/paper/2201.09012/) (2022)
- [UQuAD1.0: Development of an Urdu Question Answering Training Data for Machine Reading Comprehension](https://mlsanity.com/paper/2111.01543/) (2021)
- [A Survey on non-English Question Answering Dataset](https://mlsanity.com/paper/2112.13634/) (2021)
- [ViMMRC 2.0 -- Enhancing Machine Reading Comprehension on Vietnamese Literature Text](https://mlsanity.com/paper/2303.18162/) (2023)
- [bgGLUE: A Bulgarian General Language Understanding Evaluation Benchmark](https://mlsanity.com/paper/2306.02349/) (2023)
- [Generating Answer Candidates for Quizzes and Answer-Aware Question Generators](https://mlsanity.com/paper/2108.12898/) (2021)

## Abstract (arXiv)

Recently, reading comprehension models achieved near-human performance on large-scale datasets such as SQuAD, CoQA, MS Macro, RACE, etc. This is largely due to the release of pre-trained contextualized representations such as BERT and ELMo, which can be fine-tuned for the target task. Despite those advances and the creation of more challenging datasets, most of the work is still done for English. Here, we study the effectiveness of multilingual BERT fine-tuned on large-scale English datasets for reading comprehension (e.g., for RACE), and we apply it to Bulgarian multiple-choice reading comprehension. We propose a new dataset containing 2,221 questions from matriculation exams for twelfth grade in various subjects -history, biology, geography and philosophy-, and 412 additional questions from online quizzes in history. While the quiz authors gave no relevant context, we incorporate knowledge from Wikipedia, retrieving documents matching the combination of question + each answer option. Moreover, we experiment with different indexing and pre-training strategies. The evaluation results show accuracy of 42.23%, which is well above the baseline of 24.89%.