Targeted Knowledge Infusion To Make Conversational AI Explainable and Safe

Gaur, Manas. "Targeted Knowledge Infusion To Make Conversational AI Explainable and Safe." The Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI-23). https://ebiquity.umbc.edu/paper/html/id/1052/Targeted-Knowledge-Infusion-To-Make-Conversational-AI-Explainable-and-Safe#:~:text=Targeted%20Knowledge%20Infusion%20To%20Make%20Conversational%20AI%20Explainable%20and%20Safe,-Manas%20Gaur&text=Conversational%20Systems%20(CSys)%20represent%20practical,amount%20of%20generic%20training%20data.

Rights

This item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.

Subjects

Knowledge-infused Learning
Process Knowledge
NeuroSymbolic AI
Conversational Systems
Explainability
Safety
UMBC Ebiquity Research Group

Abstract

Conversational Systems (CSys) represent practical and tan- gible outcomes of advances in NLP and AI. CSys see continuous improvements through unsupervised training of large language models (LLMs) on a humongous amount of generic training data. However, when these CSys are suggested for use in domains like Mental Health, they fail to match the acceptable standards of clinical care, such as the clinical process in Patient Health Questionnaire (PHQ-9). The talk will present, Knowledge-infused Learning (KiL), a paradigm within NeuroSymbolic AI that focuses on making machine/deep learning models (i) learn over knowledge-enriched data, (ii) learn to follow guidelines in process-oriented tasks for safe and reasonable generation, and (iii) learn to leverage multiple contexts and stratified knowledge to yield user-level explanations (Gaur et al. 2022a). KiL established Knowledge-Intensive Language Understanding, a set of tasks for assessing safety, explainability, and conceptual flow in CSys (Sheth et al. 2021).