Non-Essential Is NEcessary: Order-agnostic Multi-hop Question Generation

Kyungho Kim, Seongmin Park, Junseo Lee, Jihwa Lee


Abstract
Existing multi-hop question generation (QG) methods treat answer-irrelevant documents as non-essential and remove them as impurities. However, this approach can create a training-inference discrepancy when impurities cannot be completely removed, which can lead to a decrease in model performance. To overcome this problem, we propose an auxiliary task, called order-agnostic, which leverages non-essential data in the training phase to create a robust model and extract the consistent embeddings in real-world inference environments. Additionally, we use a single LM to perform both ranker and generator through a prompt-based approach without applying additional external modules. Furthermore, we discover that appropriate utilization of the non-essential components can achieve a significant performance increase. Finally, experiments conducted on HotpotQA dataset achieve state-of-the-art.
Anthology ID:
2024.lrec-main.1075
Volume:
Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
Month:
May
Year:
2024
Address:
Torino, Italia
Editors:
Nicoletta Calzolari, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, Nianwen Xue
Venues:
LREC | COLING
SIG:
Publisher:
ELRA and ICCL
Note:
Pages:
12300–12306
Language:
URL:
https://aclanthology.org/2024.lrec-main.1075
DOI:
Bibkey:
Cite (ACL):
Kyungho Kim, Seongmin Park, Junseo Lee, and Jihwa Lee. 2024. Non-Essential Is NEcessary: Order-agnostic Multi-hop Question Generation. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 12300–12306, Torino, Italia. ELRA and ICCL.
Cite (Informal):
Non-Essential Is NEcessary: Order-agnostic Multi-hop Question Generation (Kim et al., LREC-COLING 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.lrec-main.1075.pdf