Speculative Decoding and Beyond: An In-Depth Survey of Techniques
This is an incremental survey paper that provides a systematic foundation for researchers and practitioners working on efficient autoregressive decoding in real-time applications.
This survey tackles the bottleneck of sequential dependencies in large-scale autoregressive models by comprehensively analyzing generation-refinement frameworks, categorizing methods based on generation strategies and refinement mechanisms, and examining deployment strategies across computing environments and applications like text, images, and speech generation.
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.