Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.25739 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912676694720512 |
|---|---|
| author | Chen, Zhi-Kai Jiang, Jun-Peng Ye, Han-Jia Zhan, De-Chuan |
| author_facet | Chen, Zhi-Kai Jiang, Jun-Peng Ye, Han-Jia Zhan, De-Chuan |
| contents | Autoregressive (AR) image generation models are capable of producing high-fidelity images but often suffer from slow inference due to their inherently sequential, token-by-token decoding process. Speculative decoding, which employs a lightweight draft model to approximate the output of a larger AR model, has shown promise in accelerating text generation without compromising quality. However, its application to image generation remains largely underexplored. The challenges stem from a significantly larger sampling space, which complicates the alignment between the draft and target model outputs, coupled with the inadequate use of the two-dimensional spatial structure inherent in images, thereby limiting the modeling of local dependencies. To overcome these challenges, we introduce Hawk, a new approach that harnesses the spatial structure of images to guide the speculative model toward more accurate and efficient predictions. Experimental results on multiple text-to-image benchmarks demonstrate a 1.71x speedup over standard AR models, while preserving both image fidelity and diversity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_25739 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Hawk: Leveraging Spatial Context for Faster Autoregressive Text-to-Image Generation Chen, Zhi-Kai Jiang, Jun-Peng Ye, Han-Jia Zhan, De-Chuan Computer Vision and Pattern Recognition Machine Learning Autoregressive (AR) image generation models are capable of producing high-fidelity images but often suffer from slow inference due to their inherently sequential, token-by-token decoding process. Speculative decoding, which employs a lightweight draft model to approximate the output of a larger AR model, has shown promise in accelerating text generation without compromising quality. However, its application to image generation remains largely underexplored. The challenges stem from a significantly larger sampling space, which complicates the alignment between the draft and target model outputs, coupled with the inadequate use of the two-dimensional spatial structure inherent in images, thereby limiting the modeling of local dependencies. To overcome these challenges, we introduce Hawk, a new approach that harnesses the spatial structure of images to guide the speculative model toward more accurate and efficient predictions. Experimental results on multiple text-to-image benchmarks demonstrate a 1.71x speedup over standard AR models, while preserving both image fidelity and diversity. |
| title | Hawk: Leveraging Spatial Context for Faster Autoregressive Text-to-Image Generation |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2510.25739 |