Review, Remask, Refine (R3): Process-Guided Block Diffusion for Text Generation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912475636563968 |
|---|---|
| author | Mounier, Nikita Idehpour, Parsa |
| author_facet | Mounier, Nikita Idehpour, Parsa |
| contents | A key challenge for iterative text generation is enabling models to efficiently identify and correct their own errors. We propose Review, Remask, Refine (R3), a relatively simple yet elegant framework that requires no additional model training and can be applied to any pre-trained masked text diffusion model (e.g., LLaDA or BD3-LM). In R3, a Process Reward Model (PRM) is utilized for the Review of intermediate generated blocks. The framework then translates these PRM scores into a Remask strategy: the lower a block's PRM score, indicating potential mistakes, the greater the proportion of tokens within that block are remasked. Finally, the model is compelled to Refine these targeted segments, focusing its efforts more intensively on specific sub-optimal parts of past generations, leading to improved final output. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_08018 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Review, Remask, Refine (R3): Process-Guided Block Diffusion for Text Generation Mounier, Nikita Idehpour, Parsa Computation and Language Machine Learning A key challenge for iterative text generation is enabling models to efficiently identify and correct their own errors. We propose Review, Remask, Refine (R3), a relatively simple yet elegant framework that requires no additional model training and can be applied to any pre-trained masked text diffusion model (e.g., LLaDA or BD3-LM). In R3, a Process Reward Model (PRM) is utilized for the Review of intermediate generated blocks. The framework then translates these PRM scores into a Remask strategy: the lower a block's PRM score, indicating potential mistakes, the greater the proportion of tokens within that block are remasked. Finally, the model is compelled to Refine these targeted segments, focusing its efforts more intensively on specific sub-optimal parts of past generations, leading to improved final output. |
| title | Review, Remask, Refine (R3): Process-Guided Block Diffusion for Text Generation |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2507.08018 |