English
Related papers

Related papers: Faster Cascades via Speculative Decoding

200 papers

Speculative decoding accelerates LLM inference by drafting multiple tokens and verifying them in parallel with the target model. However, its practical speedup is constrained by the trade-off between draft quality and drafting cost:…

Computation and Language · Computer Science 2026-05-29 Jianuo Huang , Yaojie Zhang , Qituan Zhang , Hao Lin , Hanlin Xu , Linfeng Zhang

Speculative sampling is a popular technique for accelerating inference in Large Language Models by generating candidate tokens using a fast draft model and accepting or rejecting them based on the target model's distribution. While…

Machine Learning · Computer Science 2025-07-08 Valentin De Bortoli , Alexandre Galashov , Arthur Gretton , Arnaud Doucet

The most common training pipeline for large language models includes pretraining, finetuning and aligning phases, with their respective resulting models, such as the pretrained model and the finetuned model. Finetuned and aligned models…

Computation and Language · Computer Science 2024-02-29 Lifeng Jin , Baolin Peng , Linfeng Song , Haitao Mi , Ye Tian , Dong Yu

High-performance dynamic language implementations make heavy use of speculative optimizations to achieve speeds close to statically compiled languages. These optimizations are typically performed by a just-in-time compiler that generates…

Programming Languages · Computer Science 2020-05-19 Olivier Flückiger , Gabriel Scherer , Ming-Ho Yee , Aviral Goel , Amal Ahmed , Jan Vitek

Large language models (LLMs) exhibit exceptional performance across a wide range of tasks; however, their token-by-token autoregressive generation process significantly hinders inference speed. Speculative decoding presents a promising…

Computation and Language · Computer Science 2025-03-04 Kai Lv , Honglin Guo , Qipeng Guo , Xipeng Qiu

The availability of a wide range of large language models (LLMs) embedded in various agentic systems has significantly increased the potential of model selection strategies to improve the cost-performance tradeoff. Existing strategies…

Computation and Language · Computer Science 2025-05-23 Jasper Dekoninck , Maximilian Baader , Martin Vechev

Large language models (LLMs) have achieved impressive results on multi-step mathematical reasoning, yet at the cost of high computational overhead. This challenge is particularly acute for test-time scaling methods such as parallel…

Machine Learning · Computer Science 2026-03-24 Yuanlin Chu , Bo Wang , Xiang Liu , Hong Chen , Aiwei Liu , Xuming Hu

Autoregressive decoding makes the inference of Large Language Models (LLMs) time-consuming. In this paper, we reconsider speculative sampling and derive two key observations. Firstly, autoregression at the feature (second-to-top-layer)…

Machine Learning · Computer Science 2025-03-05 Yuhui Li , Fangyun Wei , Chao Zhang , Hongyang Zhang

Speculative decoding has emerged as a promising approach to accelerate autoregressive inference in large language models (LLMs). Self-draft methods, which leverage the base LLM itself for speculation, avoid the overhead of auxiliary draft…

Computation and Language · Computer Science 2026-04-15 Zhuofan Wen , Yang Feng

Diffusion Language Models (DLMs) have recently achieved significant success due to their any-order generation capabilities. However, existing inference methods typically rely on local, immediate-step metrics such as confidence or entropy…

Computation and Language · Computer Science 2025-12-03 Kecheng Chen , Ziru Liu , Xijia Tao , Hui Liu , Xinyu Fu , Suiyun Zhang , Dandan Tu , Lingpeng Kong , Rui Liu , Haoliang Li

Large transformer-based language models have been shown to be very effective in many classification tasks. However, their computational complexity prevents their use in applications requiring the classification of a large set of candidates.…

Computation and Language · Computer Science 2020-05-08 Luca Soldaini , Alessandro Moschitti

The emergence of long-context large language models (LLMs) offers a promising alternative to traditional retrieval-augmented generation (RAG) for processing extensive documents. However, the computational overhead of long-context inference…

Computation and Language · Computer Science 2025-06-24 Guanzheng Chen , Qilong Feng , Jinjie Ni , Xin Li , Michael Qizhe Shieh

Large Language Models (LLMs) have become more prevalent in long-context applications such as interactive chatbots, document analysis, and agent workflows, but it is challenging to serve long-context requests with low latency and high…

Computation and Language · Computer Science 2025-04-03 Ranajoy Sadhukhan , Jian Chen , Zhuoming Chen , Vashisth Tiwari , Ruihang Lai , Jinyuan Shi , Ian En-Hsu Yen , Avner May , Tianqi Chen , Beidi Chen

Autoregressive Large Language Models (LLMs) have achieved impressive performance in language tasks but face two significant bottlenecks: (1) quadratic complexity in the attention module as the number of tokens increases, and (2) limited…

Computation and Language · Computer Science 2024-07-26 Haoran You , Yichao Fu , Zheng Wang , Amir Yazdanbakhsh , Yingyan Celine Lin

Speculative decoding accelerates LLMs by using a lightweight draft model to generate tokens autoregressively before verifying them in parallel with a larger target model. However, determining the optimal number of tokens to draft remains a…

Machine Learning · Computer Science 2025-11-05 Aditya Sridhar , Nish Sinnadurai , Sean Lie , Vithursan Thangarasa

Speculative decoding (SD) is a promising method for accelerating the decoding process of Large Language Models (LLMs). The efficiency of SD primarily hinges on the consistency between the draft model and the verify model. However, existing…

Computation and Language · Computer Science 2025-06-02 Longze Chen , Renke Shan , Huiming Wang , Lu Wang , Ziqiang Liu , Run Luo , Jiawei Wang , Hamid Alinejad-Rokny , Min Yang

Simultaneous translation, which starts translating each sentence after receiving only a few words in source sentence, has a vital role in many scenarios. Although the previous prefix-to-prefix framework is considered suitable for…

Computation and Language · Computer Science 2022-01-03 Zhengxin Yang

Speculative decoding is an effective and lossless method for Large Language Model (LLM) inference acceleration. It employs a smaller model to generate a draft token sequence, which is then verified by the original base model. In multi-GPU…

Machine Learning · Computer Science 2025-12-09 Yize Wu , Ke Gao , Ling Li , Yanjun Wu

This study investigates semantic uncertainty in large language model (LLM) outputs across different decoding methods, focusing on emerging techniques like speculative sampling and chain-of-thought (CoT) decoding. Through experiments on…

Computation and Language · Computer Science 2025-06-24 Darius Foodeei , Simin Fan , Martin Jaggi

Large language models (LLMs) have achieved remarkable performance across a wide range of tasks, but their increasing parameter sizes significantly slow down inference. Speculative decoding mitigates this issue by leveraging a smaller draft…

Computation and Language · Computer Science 2026-05-27 Kuan-Wei Lu , Ding-Yong Hong , Pangfeng Liu , Jan-Jan Wu