Crossing Variational Autoencoders for Answer Retrieval
This addresses answer retrieval for question-answering systems, offering an incremental improvement over existing methods.
The paper tackled the problem of answer retrieval by proposing a method that crosses variational auto-encoders to generate questions with aligned answers and answers with aligned questions, improving alignment and semantics; it outperformed the state-of-the-art method on SQuAD.
Answer retrieval is to find the most aligned answer from a large set of candidates given a question. Learning vector representations of questions/answers is the key factor. Question-answer alignment and question/answer semantics are two important signals for learning the representations. Existing methods learned semantic representations with dual encoders or dual variational auto-encoders. The semantic information was learned from language models or question-to-question (answer-to-answer) generative processes. However, the alignment and semantics were too separate to capture the aligned semantics between question and answer. In this work, we propose to cross variational auto-encoders by generating questions with aligned answers and generating answers with aligned questions. Experiments show that our method outperforms the state-of-the-art answer retrieval method on SQuAD.