A Trustworthy, Responsible and Interpretable System to Handle Chit Chat in Conversational Bots
This addresses the problem of making professional bots more engaging and appropriate for users by handling casual conversations, though it appears incremental as it combines existing techniques like classification and generation models.
The paper tackles the challenge of enabling conversational bots to handle unscripted chit-chat queries, which is difficult due to low human consensus on intent classification (77% agreement), and proposes a system that integrates hierarchical intents, sequence-to-sequence models, and interpretable components for scalable and trustworthy responses.
Most often, chat-bots are built to solve the purpose of a search engine or a human assistant: Their primary goal is to provide information to the user or help them complete a task. However, these chat-bots are incapable of responding to unscripted queries like "Hi, what's up", "What's your favourite food". Human evaluation judgments show that 4 humans come to a consensus on the intent of a given query which is from chat domain only 77% of the time, thus making it evident how non-trivial this task is. In our work, we show why it is difficult to break the chitchat space into clearly defined intents. We propose a system to handle this task in chat-bots, keeping in mind scalability, interpretability, appropriateness, trustworthiness, relevance and coverage. Our work introduces a pipeline for query understanding in chitchat using hierarchical intents as well as a way to use seq-seq auto-generation models in professional bots. We explore an interpretable model for chat domain detection and also show how various components such as adult/offensive classification, grammars/regex patterns, curated personality based responses, generic guided evasive responses and response generation models can be combined in a scalable way to solve this problem.