Graph Integrated Language Transformers for Next Action Prediction in Complex Phone Calls
This work addresses the problem of maintaining complex pipelines in dialogue managers for conversational AI, offering a more integrated approach.
The paper tackles the complexity and noise in conversational AI systems by integrating graphs into language transformers to understand relationships between utterances and actions without external dependencies, achieving higher performance in real-world phone calls.
Current Conversational AI systems employ different machine learning pipelines, as well as external knowledge sources and business logic to predict the next action. Maintaining various components in dialogue managers' pipeline adds complexity in expansion and updates, increases processing time, and causes additive noise through the pipeline that can lead to incorrect next action prediction. This paper investigates graph integration into language transformers to improve understanding the relationships between humans' utterances, previous, and next actions without the dependency on external sources or components. Experimental analyses on real calls indicate that the proposed Graph Integrated Language Transformer models can achieve higher performance compared to other production level conversational AI systems in driving interactive calls with human users in real-world settings.