TPP-SD: Accelerating Transformer Point Process Sampling with Speculative Decoding
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914104738840576 |
|---|---|
| author | Gong, Shukai Fu, Yiyang Ran, Fengyuan Kong, Quyu Zhou, Feng |
| author_facet | Gong, Shukai Fu, Yiyang Ran, Fengyuan Kong, Quyu Zhou, Feng |
| contents | We propose TPP-SD, a novel approach that accelerates Transformer temporal point process (TPP) sampling by adapting speculative decoding (SD) techniques from language models. By identifying the structural similarities between thinning algorithms for TPPs and speculative decoding for language models, we develop an efficient sampling framework that leverages a smaller draft model to generate multiple candidate events, which are then verified by the larger target model in parallel. TPP-SD maintains the same output distribution as autoregressive sampling while achieving significant acceleration. Experiments on both synthetic and real datasets demonstrate that our approach produces samples from identical distributions as standard methods, but with 2-6$\times$ speedup. Our ablation studies analyze the impact of hyperparameters such as draft length and draft model size on sampling efficiency. TPP-SD bridges the gap between powerful Transformer TPP models and the practical need for rapid sequence sampling. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_09252 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TPP-SD: Accelerating Transformer Point Process Sampling with Speculative Decoding Gong, Shukai Fu, Yiyang Ran, Fengyuan Kong, Quyu Zhou, Feng Machine Learning We propose TPP-SD, a novel approach that accelerates Transformer temporal point process (TPP) sampling by adapting speculative decoding (SD) techniques from language models. By identifying the structural similarities between thinning algorithms for TPPs and speculative decoding for language models, we develop an efficient sampling framework that leverages a smaller draft model to generate multiple candidate events, which are then verified by the larger target model in parallel. TPP-SD maintains the same output distribution as autoregressive sampling while achieving significant acceleration. Experiments on both synthetic and real datasets demonstrate that our approach produces samples from identical distributions as standard methods, but with 2-6$\times$ speedup. Our ablation studies analyze the impact of hyperparameters such as draft length and draft model size on sampling efficiency. TPP-SD bridges the gap between powerful Transformer TPP models and the practical need for rapid sequence sampling. |
| title | TPP-SD: Accelerating Transformer Point Process Sampling with Speculative Decoding |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2507.09252 |