paper-with-me

Papers

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

2023-09-15 · Jun Zhang, Jue Wang, Huan Li, Lidan Shou, Ke Chen, Gang Chen, Sharad Mehrotra

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$\times$.

📄 PDF Abstract BibTeX arXiv:2309.08168

Code (1)

dilab-zju/self-speculative-decoding 공식 구현 pytorch

Tasks

Language ModelingLanguage ModellingLarge Language Model

Similar Papers 제목 키워드 기반

Lossless Acceleration of Large Language Models with Hierarchical Drafting based on Temporal Locality in Speculative Decoding

2025-02-08 · Sukmin Cho, Sangjin Choi, Taeho Hwang, Jeongyeon Seo 외

Accelerating inference in Large Language Models (LLMs) is critical for real-time interactions, as they have been widely incorporated into real-world services. Speculative decoding, a fully algorithmic solution, has gaine…

FastEagle: Cascaded Drafting for Accelerating Speculative Decoding

2025-09-24 · Haiduo Huang, Jiangcheng Song, Wenzhe Zhao, Pengju Ren arxiv

Speculative decoding accelerates generation by drafting candidates and verifying them in parallel, yet state-of-the-art drafters (e.g., EAGLE) still require N sequential passes to propose N tokens. We present FastEagle, …

OPT-Tree: Speculative Decoding with Adaptive Draft Tree Structure

2024-06-25 · Jikai Wang, Yi Su, Juntao Li, Qingrong Xia 외

Autoregressive language models demonstrate excellent performance in various scenarios. However, the inference efficiency is limited by its one-step-one-word generation mode, which has become a pressing problem recently a…

BiTA: Bi-Directional Tuning for Lossless Acceleration in Large Language Models

2024-01-23 · Feng Lin, Hanling Yi, Hongbin Li, Yifan Yang 외

Large language models (LLMs) commonly employ autoregressive generation during inference, leading to high memory bandwidth demand and consequently extended latency. To mitigate this inefficiency, we present Bi-directional…

CAS-Spec: Cascade Adaptive Self-Speculative Decoding for On-the-Fly Lossless Inference Acceleration of LLMs

2025-10-30 · Zhiyuan Ning, Jiawei Shao, Ruge Xu, Xinfei Guo 외 arxiv

Speculative decoding has become a widely adopted as an effective technique for lossless inference acceleration when deploying large language models (LLMs). While on-the-fly self-speculative methods offer seamless integra…