Learning to Transform, Combine, and Reason in Open-domain Question Answering
| Authors | |
|---|---|
| Publication date | 2019 |
| Book title | WSDM'19 |
| Book subtitle | proceedings of the Twelfth ACM International Conference on Web Search and Data Mining : February 11-15, 2019 : Melbourne, Australia |
| ISBN (electronic) |
|
| Event | 12th ACM International Conference on Web Search and Data Mining, WSDM 2019 |
| Pages (from-to) | 681–689 |
| Publisher | New York, NY: Association for Computing Machinery |
| Organisations |
|
| Abstract |
Users seek direct answers to complex questions from large open-domain knowledge sources like the Web. Open-domain question answering has become a critical task to be solved for building systems that help address users' complex information needs. Most open-domain question answering systems use a search engine to retrieve a set of candidate documents, select one or a few of them as context, and then apply reading comprehension models to extract answers. Some questions, however, require taking a broader context into account, e.g., by considering low-ranked documents that are not immediately relevant, combining information from multiple documents, and reasoning over multiple facts from these documents to infer the answer. In this paper, we propose a model based on the Transformer architecture that is able to efficiently operate over a larger set of candidate documents by effectively combining the evidence from these documents during multiple steps of reasoning, while it is robust against noise from low-ranked non-relevant documents included in the set. We use our proposed model, called TraCRNet, on two public open-domain question answering datasets, SearchQA and Quasar-T, and achieve results that meet or exceed the state-of-the-art.
|
| Document type | Conference contribution |
| Language | English |
| Published at | https://doi.org/10.1145/3289600.3291012 |
| Downloads |
3289600.3291012
(Final published version)
|
| Permalink to this page | |
