We present a neural approach called IRNet for complex and cross-domain Text-to-SQL. IR-Net aims to address two challenges: 1) the mismatch between intents expressed in natural language (NL) and the implementation details in SQL; 2) the challenge in predicting columns caused by the large number of outof-domain words. Instead of end-to-end synthesizing a SQL query, IRNet decomposes the synthesis process into three phases. In the first phase, IRNet performs a schema linking over a question and a database schema. Then, IRNet adopts a grammar-based neural model to synthesize a SemQL query which is an intermediate representation that we design to bridge NL and SQL. Finally, IRNet deterministically infers a SQL query from the synthesized SemQL query with domain knowledge. On the challenging Text-to-SQL benchmark Spider, IRNet achieves 46.7% accuracy, obtaining 19.5% absolute improvement over previous state-of-the-art approaches. At the time of writing, IRNet achieves the first position on the Spider leaderboard.

No abstract

Most of existing studies on parsing natural language (NL) for constructing structured query language (SQL) do not consider the complex structure of database schema and the gap between NL and SQL query. In this paper, we propose a schema-aware neural network with decomposing architecture, namely HSRNet, which aims to address the complex and cross-domain Text-to-SQL generation task. The HSRNet models the relationship of the database schema with a hierarchical schema graph and employs a graph network to encode the information into sentence representation. Instead of end-to-end generation, the HSRNet decomposes the generation process into three phases. Given an input question and schema, we first choose the column candidates and generate the sketch grammar of the SQL query. Then, a detail completion module fills the details based on the column candidates and the corresponding sketch. We demonstrate the effectiveness of our hierarchical schema representation by incorporating the information into different baselines. We further show that the decomposing architecture significantly improves the performance of our model. Evaluation of Spider benchmark shows that the hierarchical schema representation and decomposing architecture improves our parser result by 14.5% and 4.3% respectively.

Artificial Intelligence (AI) personal assistant has attracted much attention from both academia and industry. Almost all existing AI personal assistants serve as service terminals to chat with human users for certain tasks. We are instead interested in building AI personal assistants for a different yet important dialog scenario, where they chat with people to fulfill specific tasks on behalf of their human users. As the personal assistants are playing a requester role, instead of a service terminal role, the conversation goal becomes delivering or requesting information according to specific user requests precisely and efficiently. The challenge for the conversation policy is that all user requests must be delivered precisely, while the challenge for the response generation is that it's generally expected for machine generated responses to cover multiple information slots, either requesting or delivering, to make the conversation efficient. In this paper, we present Table-to-Dialogue, a novel approach to address the above challenges when building a requester role AI personal assistant. We employ an encoder-decoder network to learn explicit conversation policy, which generates the corresponding information slots based on the conversation context and the user request table. We further integrate a novel Multi-Slot Constrained Bi-directional Decoder (MS-CBD) into the above encoder-decoder network, to generate machine response according to the multiple slot values and their intermediate representations from the policy decoder. Different from the existing single direction text decoder approaches, MS-CBD leverage the bi-directional context of the response when generating it to enhance the semantic coherence. The experiments shows that our approach significantly outperform the state-of-the-art conversation approaches on automatic and human evaluation metrics. INDEX TERMS Dialogue system, sequence generation, deep learning, constrained decoding, table-to-text, natural language process.