The Whole Truth and Nothing But the Truth: Faithful and Controllable Dialogue Response Generation with Dataflow Transduction and Constrained Decoding

Anonymous

Keywords: natural language generation, dialogue response generation, dialogue system, constrained decoding

Abstract: In a real-world dialogue system, generated text must satisfy several interlocking constraints: informativeness, truthfulness, and ease of control. The two predominant paradigms in language generation---neural language modeling and rule-based generation---struggle to satisfy these constraints simultaneously. We describe a hybrid architecture for dialogue response generation that combines the strengths of both paradigms. The first component of this architecture is a rule-based content selection model defined using a new formal framework called dataflow transduction, which uses declarative rules to transduce a dialogue agent’s computations (represented as dataflow graphs) into context-free grammars representing the space of contextually acceptable responses. The second component is a constrained decoding procedure that uses these grammars to constrain the output of a neural language model, which selects fluent utterances. Our experiments show that this system outperforms both rule-based and learned approaches in human evaluations of fluency, relevance, and truthfulness.

Paper Type: long

Research Area: Dialogue and Interactive Systems