paper-with-me

홈 › Papers

EAGLE-2: Faster Inference of Language Models with Dynamic Draft Trees

2024-06-24 · Yuhui Li, Fangyun Wei, Chao Zhang, Hongyang Zhang

Inference with modern Large Language Models (LLMs) is expensive and time-consuming, and speculative sampling has proven to be an effective solution. Most speculative sampling methods such as EAGLE use a static draft tree, implicitly assuming that the acceptance rate of draft tokens depends only on their position. Interestingly, we found that the acceptance rate of draft tokens is also context-dependent. In this paper, building upon EAGLE, we propose EAGLE-2, which introduces a new technique of context-aware dynamic draft tree into drafting modeling. This improvement leverages the fact that the draft model of EAGLE is well-calibrated: the confidence scores from the draft model approximate acceptance rates with small errors. We conducted extensive evaluations on three series of LLMs and six tasks, with EAGLE-2 achieving speedup ratios 3.05x-4.26x, which is 20%-40% faster than EAGLE-1. EAGLE-2 also ensures that the distribution of the generated text remains unchanged, making it a lossless acceleration algorithm.

📄 PDF Abstract BibTeX arXiv:2406.16858

Code (1)

safeailab/eagle 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Dynamic Depth Decoding: Faster Speculative Decoding for LLMs

2024-08-30 · Oscar Brown, Zhengjie Wang, Andrea Do, Nikhil Mathew 외

The acceleration of Large Language Models (LLMs) with speculative decoding provides a significant runtime improvement without any loss of accuracy. Currently, EAGLE-2 is the state-of-the-art speculative decoding method, …

SpecForge: A Flexible and Efficient Open-Source Training Framework for Speculative Decoding

2026-03-19 · Shenggui Li, Chao Wang, Yikai Zhu, Yubo Wang 외 arxiv

Large language models incur high inference latency due to sequential autoregressive decoding. Speculative decoding alleviates this bottleneck by using a lightweight draft model to propose multiple tokens for batched veri…

AdaEAGLE: Optimizing Speculative Decoding via Explicit Modeling of Adaptive Draft Structures

2024-12-25 · Situo Zhang, Hankun Wang, Da Ma, Zichen Zhu 외

Speculative Decoding (SD) is a popular lossless technique for accelerating the inference of Large Language Models (LLMs). We show that the decoding speed of SD frameworks with static draft structures can be significantly…

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, …

OWL: Overcoming Window Length-Dependence in Speculative Decoding for Long-Context Inputs

2025-10-08 · Jaeseong Lee, seung-won hwang, Aurick Qiao, Gabriele Oliaro 외 arxiv

Speculative decoding promises faster inference for large language models (LLMs), yet existing methods fail to generalize to real-world settings. Benchmarks typically assume short contexts (e.g., 2K tokens), whereas pract…