PENTATRON: PErsonalized coNText-Aware Transformer for Retrieval-based cOnversational uNderstanding

TL;DR

This paper introduces PENTATRON, a scalable, personalized, context-aware transformer system for improving conversational understanding in digital assistants by correcting entities in customer queries, significantly enhancing accuracy.

Contribution

The paper presents a novel retrieval-based entity correction system combining transformer models with personalized entity indexing for improved conversational understanding.

Findings

Up to 500.97% improvement in Exact Match metric.

Demonstrates the effectiveness of personalized, context-aware entity correction.

Establishes baselines and explores language model prompts.

Abstract

Conversational understanding is an integral part of modern intelligent devices. In a large fraction of the global traffic from customers using smart digital assistants, frictions in dialogues may be attributed to incorrect understanding of the entities in a customer's query due to factors including ambiguous mentions, mispronunciation, background noise and faulty on-device signal processing. Such errors are compounded by two common deficiencies from intelligent devices namely, (1) the device not being tailored to individual customers, and (2) the device responses being unaware of the context in the conversation session. Viewing this problem via the lens of retrieval-based search engines, we build and evaluate a scalable entity correction system, PENTATRON. The system leverages a parametric transformer-based language model to learn patterns from in-session customer-device interactions coupled with a non-parametric personalized entity index to compute the correct query, which aids downstream components in reasoning about the best response. In addition to establishing baselines and demonstrating the value of personalized and context-aware systems, we use multitasking to learn the domain of the correct entity. We also investigate the utility of language model prompts. Through extensive experiments, we show a significant upward movement of the key metric (Exact Match) by up to 500.97% (relative to the baseline).