DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Myaeng, Sung-Hyon | - |
dc.contributor.advisor | 맹성현 | - |
dc.contributor.author | Cao, Minh-Son | - |
dc.date.accessioned | 2022-04-27T19:31:56Z | - |
dc.date.available | 2022-04-27T19:31:56Z | - |
dc.date.issued | 2021 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=948436&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/296111 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전산학부, 2021.2,[iii, 18 :] | - |
dc.description.abstract | We target to improve multi-hop question answering (QA) by decomposing hard questions into simpler ones that can be answered by existing single-hop QA models. Instead of manually labeling question with decompositions or building decompositions from an external question dataset, we leverage an existing question generation (QG) model to produce high-quality dataset for training a question decomposition model, in which the decompositions are semantically relevant to the questions. Then, we adapt the previous multi-hop QA architecture utilizing decomposed sub-questions, by answering sub-questions with an existing QA model and using corresponding supporting sentences to reformulate the multi-hop questions. Experiments on HOTPOTQA show that, with equivalent amount of training data, our newly-generated decompositions help to boost the performance by 3.1 points on Ans EM score. By combining with previous training data, our model outperforms all existing question-decomposition-based models, and shows competitive results with state-of-the-art graph-based approaches. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Multi-hop Question Answering▼aQuestion Decomposition▼aGeneration▼aUnsupervised Learning▼aQuestion Answering | - |
dc.subject | 다중 홉(문단) 기반 질의 응답▼a질의 분해▼a생성▼a비지도 학습▼a질의 응답 | - |
dc.title | Generation-based question decomposition for multi-hop question answering | - |
dc.title.alternative | 다중 문단 질의응답을 위한 생성기반 질의분해 기법 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전산학부, | - |
dc.contributor.alternativeauthor | 가오 민선 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.