Discriminative Information Retrieval for Question Answering Sentence Selection

By Benjamin Van Durme, Tongfei Chen. 2017

19 citations recorded by the source corpus.

We propose a framework for discriminative IR atop linguistic features, trained to improve the recall of answer candidate passage retrieval, the initial step in text-based question answering. We formalize this as an instance of linear feature-based IR, demonstrating a 34%-43% improvement in recall for candidate triage for QA.

Subfields: Artificial Intelligence, Information Retrieval & Search, Question Answering & Reading Comprehension

Citations in the corpus

• Statute Law Information Retrieval and Entailment (2019, unknown): “QA systems normally employ an information retrieval (IR) system to produce the initial set of candidates, usually treated as a black box, bag-of-words process that selects candidate passages best overlapping with the content in the question [3].”
• UvA-DARE (Digital Academic Repository) Learning to Transform, Combine, and Reason in Open-domain Question Answering (unknown): “Different ex-ternal knowledge sources such as individual textual documents [6] and collections of documents [33] have been used for answering questions in open-domain question answering settings.”