EdiText: Controllable Coarse-to-Fine Text Editing with Diffusion Language Models
This work addresses the need for precise and scalable text editing for applications such as content moderation and sentiment adjustment, though it appears incremental by combining existing techniques.
The authors tackled the problem of controllable text editing at multiple scales by proposing EdiText, which integrates SDEdit for coarse adjustments and a self-conditioning method for fine-grained control, demonstrating robust performance in tasks like toxicity and sentiment control.
We propose EdiText, a controllable text editing method that modifies the reference text to desired attributes at various scales. We integrate an SDEdit-based editing technique that allows for broad adjustments in the degree of text editing. Additionally, we introduce a novel fine-level editing method based on self-conditioning, which allows subtle control of reference text. While being capable of editing on its own, this fine-grained method, integrated with the SDEdit approach, enables EdiText to make precise adjustments within the desired range. EdiText demonstrates its controllability to robustly adjust reference text at a broad range of levels across various tasks, including toxicity control and sentiment control.