A Unified Approach to Entity-Centric Context Tracking in Social Conversations

TL;DR

This paper introduces an end-to-end entity-centric context tracking framework for social conversations, along with a new large-scale dataset and a neural network architecture, advancing the efficiency and accuracy of tracking entities in dialogue.

Contribution

It proposes a unified, end-to-end model for entity context tracking, releases a comprehensive dataset, and compares its approach with existing methods.

Findings

The neural network outperforms state-of-the-art approaches on key subtasks.

The framework efficiently handles long conversations with dynamic entity updates.

The dataset enables extensive evaluation of context tracking methods.

Abstract

In human-human conversations, Context Tracking deals with identifying important entities and keeping track of their properties and relationships. This is a challenging problem that encompasses several subtasks such as slot tagging, coreference resolution, resolving plural mentions and entity linking. We approach this problem as an end-to-end modeling task where the conversational context is represented by an entity repository containing the entity references mentioned so far, their properties and the relationships between them. The repository is updated turn-by-turn, thus making training and inference computationally efficient even for long conversations. This paper lays the groundwork for an investigation of this framework in two ways. First, we release Contrack, a large scale human-human conversation corpus for context tracking with people and location annotations. It contains over 7000 conversations with an average of 11.8 turns, 5.8 entities and 15.2 references per conversation. Second, we open-source a neural network architecture for context tracking. Finally we compare this network to state-of-the-art approaches for the subtasks it subsumes and report results on the involved tradeoffs.