Interactive Question Answering and Content Delivery
Interactions with current AI agents (e.g., Amazon Alexa, Google Home, Apple Siri) are limited to single-turn simple tasks such as asking about the weather, listening to a song, or telling a joke. What they are currently lacking is a more in-depth multi-turn conversation on wider domains (e.g., talking about the news) that entail follow-up questions, retrieving information from knowledge bases (e.g., WikiData), texts (e.g., news articles) and performing common-sense reasoning (e.g., if-then clauses).
This project proposes to build such models based on recent advances in Deep Learning, Natural Language Processing and Conversational AI. Starting from existing conversational agents that answer questions of users based on a Wikipedia article (Redy et al., 2019), we will explore the idea of delivering content to the user (e.g., telling the news) crucially being able to respond to questions addressed within the text, or found in other texts (Fan et al., 2019), knowledge bases (Moon et al., 2019) or common-sense knowledge graphs (Sap et al., 2019). In this way, the model will also be able to create a more interpretable and transparent output, thus instilling trust to the user.
The challenges we will explore in this project include neural language generation, long text encoding, knowledge base representation, and rudimentary reasoning over knowledge graphs. All of these topics are are at the epicentre of current research in NLP and are desirable features for Conversational AI companies such as ALANA AI.
- Moon et al., ACL 2019. OpenDialKG: Explainable Conversational Reasoning with Attention-based Walks over Knowledge Graphs (https://www.aclweb.org/anthology/P19-1081/)
- Redy et al., TACL 2019. CoQA: A Conversational Question Answering Challenge (https://www.aclweb.org/anthology/P19-1081/)
- Fan et al., EMNLP 2019. Using Local Knowledge Graph Construction to Scale Seq2Seq Models to Multi-Document Inputs (https://www.aclweb.org/anthology/D19-1428/)
- Sap etl al., AAAI 2019. ATOMIC: An Atlas of Machine Commonsense for If-Then Reasoning (https://arxiv.org/abs/1811.00146)