Multi-Domain Dialogue State Tracking based on State Graph
This work addresses the challenge of extracting accurate dialogue states in multi-domain conversations, which is crucial for improving virtual assistants and chatbots, representing an incremental advancement over prior methods.
The paper tackles the problem of multi-domain Dialogue State Tracking with open vocabulary by constructing a dialogue state graph to properly connect domains, slots, and values, avoiding spurious attention connections in Transformers. It achieves a new state of the art on the task, outperforming existing open-vocabulary DST approaches while remaining efficient.
We investigate the problem of multi-domain Dialogue State Tracking (DST) with open vocabulary, which aims to extract the state from the dialogue. Existing approaches usually concatenate previous dialogue state with dialogue history as the input to a bi-directional Transformer encoder. They rely on the self-attention mechanism of Transformer to connect tokens in them. However, attention may be paid to spurious connections, leading to wrong inference. In this paper, we propose to construct a dialogue state graph in which domains, slots and values from the previous dialogue state are connected properly. Through training, the graph node and edge embeddings can encode co-occurrence relations between domain-domain, slot-slot and domain-slot, reflecting the strong transition paths in general dialogue. The state graph, encoded with relational-GCN, is fused into the Transformer encoder. Experimental results show that our approach achieves a new state of the art on the task while remaining efficient. It outperforms existing open-vocabulary DST approaches.