Incorporating Stylistic Lexical Preferences in Generative Language Models
This addresses the need for controllable text generation in applications like personalized content creation, though it appears incremental as it builds on existing transformer-based models and reinforcement learning frameworks.
The authors tackled the problem of language models being unable to emulate specific target styles by developing a reinforcement learning approach that incorporates continuous multi-dimensional lexical preferences of an author, resulting in generated text that distinctively aligns with the target author's lexical style.
While recent advances in language modeling have resulted in powerful generation models, their generation style remains implicitly dependent on the training data and can not emulate a specific target style. Leveraging the generative capabilities of a transformer-based language models, we present an approach to induce certain target-author attributes by incorporating continuous multi-dimensional lexical preferences of an author into generative language models. We introduce rewarding strategies in a reinforcement learning framework that encourages the use of words across multiple categorical dimensions, to varying extents. Our experiments demonstrate that the proposed approach can generate text that distinctively aligns with a given target author's lexical style. We conduct quantitative and qualitative comparisons with competitive and relevant baselines to illustrate the benefits of the proposed approach.