CoCon: A Self-Supervised Approach for Controlled Text Generation

Abstract

Pretrained Transformer-based language models (LMs) display remarkable natural language generation capabilities. With their immense potential, controlling text generation of such LMs is getting attention. While there are studies that seek to control high-level attributes (such as sentiment and topic) of generated text, there is still a lack of more precise control over its content at the word- and phrase-level. Here, we propose Content-Conditioner (CoCon) to control an LM's output text with a content input, at a fine-grained level. In our self-supervised approach, the CoCon block learns to help the LM complete a partially-observed text sequence by conditioning with content inputs that are withheld from the LM. Through experiments, we show that CoCon can naturally incorporate target content into generated texts and control high-level text attributes in a zero-shot manner.

ICLR 2021 Camera-Ready

References in corpus (5)

• Language Models are Few-Shot Learners
• Synthesizer: Rethinking Self-Attention in Transformer Models
• Plug and Play Language Models: A Simple Approach to Controlled Text Generation
• Controlling Output Length in Neural Encoder-Decoders
• What makes a good conversation? How controllable attributes affect human judgments

Cited by in corpus (2)

• Neural Rule-Execution Tracking Machine For Transformer-Based Text Generation
• Plug-and-Blend: A Framework for Controllable Story Generation with Blended Control Codes