English

Speculative Decoding and Beyond: An In-Depth Survey of Techniques

Computation and Language 2025-10-09 v4

Abstract

Sequential dependencies present a fundamental bottleneck in deploying large-scale autoregressive models, particularly for real-time applications. While traditional optimization approaches like pruning and quantization often compromise model quality, recent advances in generation-refinement frameworks demonstrate that this trade-off can be significantly mitigated. This survey presents a comprehensive taxonomy of generation-refinement frameworks, analyzing methods across autoregressive sequence tasks. We categorize methods based on their generation strategies (from simple n-gram prediction to sophisticated draft models) and refinement mechanisms (including single-pass verification and iterative approaches). Through systematic analysis of both algorithmic innovations and system-level implementations, we examine deployment strategies across computing environments and explore applications spanning text, images, and speech generation. This systematic examination of both theoretical frameworks and practical implementations provides a foundation for future research in efficient autoregressive decoding.

Keywords

Cite

@article{arxiv.2502.19732,
  title  = {Speculative Decoding and Beyond: An In-Depth Survey of Techniques},
  author = {Yunhai Hu and Zining Liu and Zhenyuan Dong and Tianfan Peng and Bradley McDanel and Sai Qian Zhang},
  journal= {arXiv preprint arXiv:2502.19732},
  year   = {2025}
}
R2 v1 2026-06-28T21:59:36.402Z