Seen to Unseen: Exploring Compositional Generalization of Multi-Attribute Controllable Dialogue Generation
About
Existing controllable dialogue generation work focuses on the single-attribute control and lacks generalization capability to out-of-distribution multiple attribute combinations. In this paper, we explore the compositional generalization for multi-attribute controllable dialogue generation where a model can learn from seen attribute values and generalize to unseen combinations. We propose a prompt-based disentangled controllable dialogue generation model, DCG. It learns attribute concept composition by generating attribute-oriented prompt vectors and uses a disentanglement loss to disentangle different attributes for better generalization. Besides, we design a unified reference-free evaluation framework for multiple attributes with different levels of granularities. Experiment results on two benchmarks prove the effectiveness of our method and the evaluation metric.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Attribute-Controlled Dialogue Generation | DailyDialog-CG (test) | Emotion Accuracy (E-ACC)70.66 | 12 | |
| Multi-Aspect Controllable Text Generation | Fyelp ACD CompMCTG | Acomp75.72 | 12 | |
| Multi-Aspect Controllable Text Generation | Fyelp CompMCTG (Hold-Out) | Acomp80.61 | 12 | |
| Multi-Aspect Controllable Text Generation | CompMCTG Overall Summary Average 1.0 | Aavg Score76.49 | 10 | |
| Multi-attribute Conditional Text Generation | CompMCTG Compositional Few-Shot 1.0 (test) | Accuracy59.27 | 10 | |
| Multi-attribute Controlled Text Generation | CompM-CTG (Hold-Out) | Dist-3 (i.d.)0.694 | 10 | |
| Multi-attribute Controlled Text Generation | CompM-CTG Average | Dist-3 Average0.691 | 10 | |
| Multi-Constraint Text Generation | CompMCTG Average 1.0 | Relevance (avg)3.79 | 10 | |
| Multi-Aspect Controllable Text Generation | CompMCTG 1.0 (Original) | Aid Score79.93 | 10 | |
| Multi-Aspect Controllable Text Generation | CompMCTG 1.0 (Hold-Out) | Aid Score79.72 | 10 |