paper-with-me

Papers

Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding

2024-01-15 · Heming Xia, Zhe Yang, Qingxiu Dong, Peiyi Wang, Yongqi Li, Tao Ge, Tianyu Liu, Wenjie Li, Zhifang Sui

To mitigate the high inference latency stemming from autoregressive decoding in Large Language Models (LLMs), Speculative Decoding has emerged as a novel decoding paradigm for LLM inference. In each decoding step, this method first drafts several future tokens efficiently and then verifies them in parallel. Unlike autoregressive decoding, Speculative Decoding facilitates the simultaneous decoding of multiple tokens per step, thereby accelerating inference. This paper presents a comprehensive overview and analysis of this promising decoding paradigm. We begin by providing a formal definition and formulation of Speculative Decoding. Then, we organize in-depth discussions on its key facets, such as drafter selection and verification strategies. Furthermore, we present a comparative analysis of leading methods under third-party testing environments. We aim for this work to serve as a catalyst for further research on Speculative Decoding, ultimately contributing to more efficient LLM inference.

📄 PDF Abstract BibTeX arXiv:2401.07851

Code (2)

hemingkx/speculativedecodingpapers 공식 구현 pytorch
hemingkx/Spec-Bench pytorch

Tasks

Language ModelingLanguage ModellingLarge Language Model

Similar Papers 제목 키워드 기반

Encoder-Decoder or Decoder-Only? Revisiting Encoder-Decoder Large Language Model

2025-10-30 · Biao Zhang, Yong Cheng, Siamak Shakeri, Xinyi Wang 외 arxiv

Recent large language model (LLM) research has undergone an architectural shift from encoder-decoder modeling to nowadays the dominant decoder-only modeling. This rapid transition, however, comes without a rigorous compa…

Unlocking Parameter-Efficient Fine-Tuning for Low-Resource Language Translation

2024-04-05 · Tong Su, Xin Peng, Sarubi Thillainathan, David Guzmán 외

Parameter-efficient fine-tuning (PEFT) methods are increasingly vital in adapting large-scale pre-trained language models for diverse tasks, offering a balance between adaptability and computational efficiency. They are …

Computational EfficiencyMachine TranslationNMTparameter-efficient fine-tuning+1

Breaking the Language Barrier: Can Direct Inference Outperform Pre-Translation in Multilingual LLM Applications?

2024-03-04 · Yotam Intrator, Matan Halfon, Roman Goldenberg, Reut Tsarfaty 외

Large language models hold significant promise in multilingual applications. However, inherent biases stemming from predominantly English-centric pre-training have led to the widespread practice of pre-translation, i.e.,…

Translation

Efficient Paths and Dense Rewards: Probabilistic Flow Reasoning for Large Language Models

2026-01-14 · Yan Liu, Feng Zhang, Zhanyu Ma, Jun Xu 외 arxiv

High-quality chain-of-thought has demonstrated strong potential for unlocking the reasoning capabilities of large language models. However, current paradigms typically treat the reasoning process as an indivisible sequen…

Reinforcement Learning

Ideas in Inference-time Scaling can Benefit Generative Pre-training Algorithms

2025-03-10 · Jiaming Song, Linqi Zhou

Recent years have seen significant advancements in foundation models through generative pre-training, yet algorithmic innovation in this space has largely stagnated around autoregressive models for discrete signals and d…