Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

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.

Weihao Zeng, Lulu Zhao, Keqing He, Ruotong Geng, Jingang Wang, Wei Wu, Weiran Xu• 2023

Related benchmarks

TaskDatasetResultRank
Attribute-Controlled Dialogue GenerationDailyDialog-CG (test)
Emotion Accuracy (E-ACC)70.66
12
Multi-Aspect Controllable Text GenerationFyelp ACD CompMCTG
Acomp75.72
12
Multi-Aspect Controllable Text GenerationFyelp CompMCTG (Hold-Out)
Acomp80.61
12
Multi-Aspect Controllable Text GenerationCompMCTG Overall Summary Average 1.0
Aavg Score76.49
10
Multi-attribute Conditional Text GenerationCompMCTG Compositional Few-Shot 1.0 (test)
Accuracy59.27
10
Multi-attribute Controlled Text GenerationCompM-CTG (Hold-Out)
Dist-3 (i.d.)0.694
10
Multi-attribute Controlled Text GenerationCompM-CTG Average
Dist-3 Average0.691
10
Multi-Constraint Text GenerationCompMCTG Average 1.0
Relevance (avg)3.79
10
Multi-Aspect Controllable Text GenerationCompMCTG 1.0 (Original)
Aid Score79.93
10
Multi-Aspect Controllable Text GenerationCompMCTG 1.0 (Hold-Out)
Aid Score79.72
10
Showing 10 of 18 rows

Other info

Code

Follow for update