RSA-Control: A Pragmatics-Grounded Lightweight Controllable Text Generation Framework
Abstract: Despite significant advancements in natural language generation, controlling language models to produce texts with desired attributes remains a formidable challenge. In this work, we introduce RSA-Control, a training-free controllable text generation framework grounded in pragmatics. RSA-Control directs the generation process by recursively reasoning between imaginary speakers and listeners, enhancing the likelihood that target attributes are correctly interpreted by listeners amidst distractors. Additionally, we introduce a self-adjustable rationality parameter, which allows for automatic adjustment of control strength based on context. Our experiments, conducted with two task types and two types of language models, demonstrate that RSA-Control achieves strong attribute control while maintaining language fluency and content consistency.
Paper Type: Long
Research Area: Generation
Research Area Keywords: inference methods, bias/toxicity
Contribution Types: NLP engineering experiment
Languages Studied: English
Submission Number: 1403
Loading