Pre-train and Plug-in: Flexible Conditional Text Generation with Variational Auto-Encoders
This addresses the inefficiency of retraining for emerging conditions in conditional text generation, which is incremental but beneficial for real-world applications.
The paper tackles the problem of conditional text generation models requiring full retraining when new conditions are added, by proposing a framework that decouples text generation from condition representation, enabling efficient plug-in updates. Experiments show PPVAE achieves better conditionality and diversity with less training effort compared to existing methods.
Conditional Text Generation has drawn much attention as a topic of Natural Language Generation (NLG) which provides the possibility for humans to control the properties of generated contents. Current conditional generation models cannot handle emerging conditions due to their joint end-to-end learning fashion. When a new condition added, these techniques require full retraining. In this paper, we present a new framework named Pre-train and Plug-in Variational Auto-Encoder (PPVAE) towards flexible conditional text generation. PPVAE decouples the text generation module from the condition representation module to allow "one-to-many" conditional generation. When a fresh condition emerges, only a lightweight network needs to be trained and works as a plug-in for PPVAE, which is efficient and desirable for real-world applications. Extensive experiments demonstrate the superiority of PPVAE against the existing alternatives with better conditionality and diversity but less training effort.