TY - CHAP

T1 - Low Resource Question Answering: An Amharic Benchmarking Dataset

T2 - 5th Workshop on Resources for African Indigenous Languages - RAIL 2024

AU - Taffa, Tilahun Abedissa

AU - Assabie, Yaregal

AU - Usbeck, Ricardo

N1 - Conference code: 5

PY - 2024

Y1 - 2024

N2 - Question Answering (QA) systems return concise answers or answer lists based on natural language text, which uses a given context document. Many resources go into curating QA datasets to advance the development of robust QA models. There is a surge in QA datasets for languages such as English; this is different for low-resource languages like Amharic. Indeed, there is no published or publicly available Amharic QA dataset. Hence, to foster further research in low-resource QA, we present the first publicly available benchmarking Amharic Question Answering Dataset (Amh-QuAD). We crowdsource 2,628 question-answer pairs from over 378 Amharic Wikipedia articles. Using the training set, we fine-tune an XLM-R-based language model and introduce a new reader model. Leveraging our newly fine-tuned reader run a baseline model to spark open-domain Amharic QA research interest. The best-performing baseline QA achieves an F-score of 80.3 and 81.34 in retriever-reader and reading comprehension settings.

AB - Question Answering (QA) systems return concise answers or answer lists based on natural language text, which uses a given context document. Many resources go into curating QA datasets to advance the development of robust QA models. There is a surge in QA datasets for languages such as English; this is different for low-resource languages like Amharic. Indeed, there is no published or publicly available Amharic QA dataset. Hence, to foster further research in low-resource QA, we present the first publicly available benchmarking Amharic Question Answering Dataset (Amh-QuAD). We crowdsource 2,628 question-answer pairs from over 378 Amharic Wikipedia articles. Using the training set, we fine-tune an XLM-R-based language model and introduce a new reader model. Leveraging our newly fine-tuned reader run a baseline model to spark open-domain Amharic QA research interest. The best-performing baseline QA achieves an F-score of 80.3 and 81.34 in retriever-reader and reading comprehension settings.

KW - Amh-QuAD

KW - Amharic Question Answering Dataset

KW - Amharic Reading Comprehension

KW - Low Resource Question Answering

KW - Informatics

UR - http://www.scopus.com/inward/record.url?scp=85195211713&partnerID=8YFLogxK

UR - https://aclanthology.org/2024.rail-1.0.pdf

UR - https://aclanthology.org/events/coling-2024/#2024rail-1

UR - https://www.mendeley.com/catalogue/3de6d397-698d-3f23-bb7f-847fca82ca94/

M3 - Article in conference proceedings

AN - SCOPUS:85195211713

SN - 9782493814401

T3 - LREC proceedings

SP - 124

EP - 132

BT - The Fifth Workshop on Resources for African Indigenous Languages @LREC-COLING-2024 (RAIL)

A2 - Mabuya, Rooweither

A2 - Matfunjwa, Muzi

A2 - Setaka, Mmasibidi

A2 - van Zaanen, Menno

PB - European Language Resources Association (ELRA)

CY - Paris

Y2 - 25 May 2024

ER -