Details
| Original language | English |
|---|---|
| Title of host publication | 41st International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2018 |
| Subtitle of host publication | The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval |
| Pages | 325-334 |
| Number of pages | 10 |
| ISBN (electronic) | 978-1-4503-5657-2 |
| Publication status | Published - 27 Jun 2018 |
Abstract
Attention based neural network models have been successfully applied in answer selection, which is an important subtask of question answering (QA). These models often represent a question by a single vector and find its corresponding matches by attending to candidate answers. However, questions and answers might be related to each other in complicated ways which cannot be captured by single-vector representations. In this paper, we propose Multihop Attention Networks (MAN) which aim to uncover these complex relations for ranking question and answer pairs. Unlike previous models, we do not collapse the question into a single vector, instead we use multiple vectors which focus on different parts of the question for its overall semantic representation and apply multiple steps of attention to learn representations for the candidate answers. For each attention step, in addition to common attention mechanisms, we adopt sequential attention which utilizes context information for computing context-aware attention weights. Via extensive experiments, we show that MAN outperforms state-of-the-art approaches on popular benchmark QA datasets. Empirical studies confirm the effectiveness of sequential attention over other attention mechanisms.
Keywords
- Answer selection, non-factoid QA, representation learning, attention mechanism, Representation learning, Attention mechanism, Non-factoid qa
ASJC Scopus subject areas
- Computer Science(all)
- Software
- Computer Science(all)
- Information Systems
- Computer Science(all)
- Computer Graphics and Computer-Aided Design
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
41st International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2018: The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval. 2018. p. 325-334.
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Multihop Attention Networks for Question Answer Matching
AU - Nam Khanh Tran,
AU - Niederee, Claudia
N1 - Publisher Copyright: © 2018 ACM.
PY - 2018/6/27
Y1 - 2018/6/27
N2 - Attention based neural network models have been successfully applied in answer selection, which is an important subtask of question answering (QA). These models often represent a question by a single vector and find its corresponding matches by attending to candidate answers. However, questions and answers might be related to each other in complicated ways which cannot be captured by single-vector representations. In this paper, we propose Multihop Attention Networks (MAN) which aim to uncover these complex relations for ranking question and answer pairs. Unlike previous models, we do not collapse the question into a single vector, instead we use multiple vectors which focus on different parts of the question for its overall semantic representation and apply multiple steps of attention to learn representations for the candidate answers. For each attention step, in addition to common attention mechanisms, we adopt sequential attention which utilizes context information for computing context-aware attention weights. Via extensive experiments, we show that MAN outperforms state-of-the-art approaches on popular benchmark QA datasets. Empirical studies confirm the effectiveness of sequential attention over other attention mechanisms.
AB - Attention based neural network models have been successfully applied in answer selection, which is an important subtask of question answering (QA). These models often represent a question by a single vector and find its corresponding matches by attending to candidate answers. However, questions and answers might be related to each other in complicated ways which cannot be captured by single-vector representations. In this paper, we propose Multihop Attention Networks (MAN) which aim to uncover these complex relations for ranking question and answer pairs. Unlike previous models, we do not collapse the question into a single vector, instead we use multiple vectors which focus on different parts of the question for its overall semantic representation and apply multiple steps of attention to learn representations for the candidate answers. For each attention step, in addition to common attention mechanisms, we adopt sequential attention which utilizes context information for computing context-aware attention weights. Via extensive experiments, we show that MAN outperforms state-of-the-art approaches on popular benchmark QA datasets. Empirical studies confirm the effectiveness of sequential attention over other attention mechanisms.
KW - Answer selection
KW - non-factoid QA
KW - representation learning
KW - attention mechanism
KW - Representation learning
KW - Attention mechanism
KW - Non-factoid qa
UR - http://www.scopus.com/inward/record.url?scp=85051505288&partnerID=8YFLogxK
U2 - 10.1145/3209978.3210009
DO - 10.1145/3209978.3210009
M3 - Conference contribution
SP - 325
EP - 334
BT - 41st International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2018
ER -