Learning Symmetric Collaborative Dialogue Agents with Dynamic Knowledge Graph Embeddings
This addresses the challenge of open-ended dialogue states in collaborative settings, but it is incremental as it builds on existing neural and rule-based models.
The paper tackled the problem of symmetric collaborative dialogue where two agents with private knowledge must communicate to achieve a common goal, and the result was a neural model with dynamic knowledge graph embeddings that outperformed baselines in effectiveness and human-likeness.
We study a symmetric collaborative dialogue setting in which two agents, each with private knowledge, must strategically communicate to achieve a common goal. The open-ended dialogue state in this setting poses new challenges for existing dialogue systems. We collected a dataset of 11K human-human dialogues, which exhibits interesting lexical, semantic, and strategic elements. To model both structured knowledge and unstructured language, we propose a neural model with dynamic knowledge graph embeddings that evolve as the dialogue progresses. Automatic and human evaluations show that our model is both more effective at achieving the goal and more human-like than baseline neural and rule-based models.