Learning Representations for Zero-Shot Retrieval over Structured Data
This addresses a bottleneck in question-answering over structured data for applications like chatbots, though it is incremental as it adapts existing retrieval methods to a new context.
The paper tackles the problem of retrieving the correct structured table from a large pool for a given query in question-answering systems, proposing an architecture that achieves this without prior table knowledge.
Large Scale Question-Answering systems today are widely used in downstream applications such as chatbots and conversational dialogue agents. Typically, such systems consist of an Answer Passage retrieval layer coupled with Machine Comprehension models trained on natural language query-passage pairs. Recent studies have explored Question Answering over structured data sources such as web-tables and relational databases. However, architectures such as Seq2SQL assume the correct table a priori which is input to the model along with the free text question. Our proposed method, analogues to a passage retrieval model in traditional Question-Answering systems, describes an architecture to discern the correct table pertaining to a given query from amongst a large pool of candidate tables.