# Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding

arXiv: https://arxiv.org/abs/2309.08168 · HTML: https://mlsanity.com/paper/2309.08168/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Placed in Speculative decoding, one of 12 directions in Efficient LLM inference and serving.
- First posted on arXiv 15 Sep 2023.

## Placed in

- Efficient LLM inference and serving / [Speculative decoding](https://mlsanity.com/inference/speculative-decoding/)

## Built on by (most cited first, 8 of 8)

- [A Survey of Resource-efficient LLM and Multimodal Foundation Models](https://mlsanity.com/paper/2401.08092/) (2024)
- [How Far Are We From AGI: Are LLMs All We Need?](https://mlsanity.com/paper/2405.10313/) (2024)
- [A Survey on Parallel Text Generation: From Parallel Decoding to Diffusion Language Models](https://mlsanity.com/paper/2508.08712/) (2025)
- [Efficient Reasoning for LLMs through Speculative Chain-of-Thought](https://mlsanity.com/paper/2504.19095/) (2025)
- [Towards AI Search Paradigm](https://mlsanity.com/paper/2506.17188/) (2025)
- [Less is More: Improving LLM Reasoning with Minimal Test-Time Intervention](https://mlsanity.com/paper/2510.13940/) (2025)
- [Thinking with Drafts: Speculative Temporal Reasoning for Efficient Long Video Understanding](https://mlsanity.com/paper/2512.00805/) (2025)
- [Nucleus Speculative Decoding: Plausibility-Aware Verification Beyond Exact Distribution](https://mlsanity.com/paper/2610.07822/) (2026)

## Abstract (arXiv)

We present a novel inference scheme, self-speculative decoding, for accelerating Large Language Models (LLMs) without the need for an auxiliary model. This approach is characterized by a two-stage process: drafting and verification. The drafting stage generates draft tokens at a slightly lower quality but more quickly, which is achieved by selectively skipping certain intermediate layers during drafting. Subsequently, the verification stage employs the original LLM to validate those draft output tokens in one forward pass. This process ensures the final output remains identical to that produced by the unaltered LLM. Moreover, the proposed method requires no additional neural network training and no extra memory footprint, making it a plug-and-play and cost-effective solution for inference acceleration. Benchmarks with LLaMA-2 and its variants demonstrated a speedup up to 1.99×.