Interpretation of Natural Language Rules in Conversational Machine Reading
Being able to answer questions is the main ability of an AI assistant. The required knowledge for answering questions often comes from unstructured data, i.e. text. Being able to read a text and answer questions related to it is referred to as Machine Reading. Most work in machine reading focuses on question answering problems where the answer is directly expressed in the text to read. However, many real-world question answering problems require the reading of text not because it contains the literal answer, but because it contains a recipe to derive an answer together with the reader's background knowledge. One example is the task of interpreting regulations to answer "Can I...?" or "Do I have to...?" questions such as "I am working in Canada. Do I have to carry on paying UK National Insurance?" after reading a UK government website about this topic. This task requires both the interpretation of rules and the application of background knowledge. It is further complicated due to the fact that, in practice, most questions are underspecified, and a human assistant will regularly have to ask clarification questions such as "How long have you been working abroad?" when the answer cannot be directly derived from the question and text. In this talk, I will give a brief overview of different types of machine reading (Question Answering) tasks and datasets. I will then introduce ShARC, a conversational machine reading dataset and task that aims to address the scenarios explained above.
I have a degree in software engineering from Ferdowsi University in Iran and then Kings College London. After working as a software engineer in finance sector for over 4 years, I did a PhD in Natural Language Processing under the supervision of Sebastian Riedel at UCL. My PhD topic was on predicting attributes of city neighborhoods using social media text. I then worked as a research scientist for Bloomsbury AI doing research on creating an AI assistant specifically to answer questions from text containing rules such as legislations and compliance. Currently, I work as a research scientist at Facebook on integrity issues with a focus on identifying false claims.