Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.20724 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915692909953024 |
|---|---|
| author | Christoforos, Alexandros Davis, Chadbourne |
| author_facet | Christoforos, Alexandros Davis, Chadbourne |
| contents | Diffusion based approaches to long form text generation suffer from prohibitive computational cost and memory overhead as sequence length increases. We introduce SA-DiffuSeq, a diffusion framework that integrates sparse attention to fundamentally improve scalability for long document modeling. By selectively allocating attention within the diffusion process, SA-DiffuSeq significantly reduces computational complexity while maintaining semantic coherence and generation quality. A key component of our method is a soft absorbing state tailored to sparse attention dynamics, which stabilizes diffusion trajectories and accelerates sequence reconstruction. This design improves sampling efficiency and enhances precision in long range dependency modeling. Extensive experiments demonstrate that SA-DiffuSeq consistently surpasses state of the art diffusion baselines in both training efficiency and sampling speed, with especially strong gains on extended sequences. These properties make SA-DiffuSeq well suited for demanding long form applications such as scientific writing, large scale code generation, and multi turn long context dialogue. Overall, our results indicate that incorporating structured sparsity into diffusion models is a promising direction for efficient and expressive long text generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_20724 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SA-DiffuSeq: Addressing Computational and Scalability Challenges in Long-Document Generation with Sparse Attention Christoforos, Alexandros Davis, Chadbourne Computation and Language Artificial Intelligence Diffusion based approaches to long form text generation suffer from prohibitive computational cost and memory overhead as sequence length increases. We introduce SA-DiffuSeq, a diffusion framework that integrates sparse attention to fundamentally improve scalability for long document modeling. By selectively allocating attention within the diffusion process, SA-DiffuSeq significantly reduces computational complexity while maintaining semantic coherence and generation quality. A key component of our method is a soft absorbing state tailored to sparse attention dynamics, which stabilizes diffusion trajectories and accelerates sequence reconstruction. This design improves sampling efficiency and enhances precision in long range dependency modeling. Extensive experiments demonstrate that SA-DiffuSeq consistently surpasses state of the art diffusion baselines in both training efficiency and sampling speed, with especially strong gains on extended sequences. These properties make SA-DiffuSeq well suited for demanding long form applications such as scientific writing, large scale code generation, and multi turn long context dialogue. Overall, our results indicate that incorporating structured sparsity into diffusion models is a promising direction for efficient and expressive long text generation. |
| title | SA-DiffuSeq: Addressing Computational and Scalability Challenges in Long-Document Generation with Sparse Attention |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2512.20724 |