English
Related papers

Related papers: Accelerating Diffusion LLM Inference via Local Det…

200 papers

Diffusion Language Models (DLMs) are often advertised as enabling parallel token generation, yet practical fast DLMs frequently converge to left-to-right, autoregressive (AR)-like decoding dynamics. In contrast, genuinely non-AR generation…

Computation and Language · Computer Science 2026-03-02 Pengxiang Li , Dilxat Muhtar , Tianlong Chen , Lu Yin , Shiwei Liu

While Diffusion Language Models (DLMs) are theoretically well-suited for iterative refinement due to their non-causal structure, they often fail to reliably revise incorrect tokens in practice. The key challenge lies in the model's…

Machine Learning · Computer Science 2026-01-30 Shuibai Zhang , Fred Zhangzhi Peng , Yiheng Zhang , Jin Pan , Grigorios G. Chrysos

Diffusion large language models (dLLMs) offer faster generation than autoregressive models while maintaining comparable quality, but existing watermarking methods fail on them due to their non-sequential decoding. Unlike autoregressive…

Machine Learning · Computer Science 2025-10-06 Linyu Wu , Linhao Zhong , Wenjie Qu , Yuexin Li , Yue Liu , Shengfang Zhai , Chunhua Shen , Jiaheng Zhang

Fine-tuning large language models (LLMs) for recommendation in a generative manner has delivered promising results, but encounters significant inference overhead due to autoregressive decoding in the language space. This work explores…

Information Retrieval · Computer Science 2025-09-16 Chengbing Wang , Yang Zhang , Zhicheng Wang , Tianhao Shi , Keqin Bao , Fuli Feng , Tat-Seng Chua

This chapter explores advancements in decoding strategies for large language models (LLMs), focusing on enhancing the Locally Typical Sampling (LTS) algorithm. Traditional decoding methods, such as top-k and nucleus sampling, often struggle…

Computation and Language · Computer Science 2025-06-12 Jaydip Sen , Saptarshi Sengupta , Subhasis Dasgupta

Large language model (LLM) serving is fundamentally limited by inefficient hardware utilization. Autoregressive (AR) decoding underutilizes GPUs due to its strictly sequential execution, while diffusion LLMs (DLLMs) improve throughput by…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-26 Chiyue Wei , Cong Guo , Bowen Duan , Junyao Zhang , Haoxuan Shan , Yifei Wang , Yangjie Zhou , Hai "Helen" Li , Danyang Zhuo , Yiran Chen

Diffusion large language models (dLLMs) have recently gained significant attention due to their inherent support for parallel decoding. Building on this paradigm, Mixture-of-Experts (MoE) dLLMs with autoregressive (AR) initialization have…

Computation and Language · Computer Science 2026-05-25 Linye Wei , Zixiang Luo , Pingzhi Tang , Meng Li

Diffusion large language models (dLLMs) are emerging as promising alternatives to autoregressive (AR) LLMs. Recently, this paradigm has been extended to multimodal tasks, leading to the development of diffusion multimodal large language…

Artificial Intelligence · Computer Science 2026-04-08 Keuntae Kim , Mingyu Kang , Yong Suk Choi

Large Language Models (LLMs) for complex reasoning is often hindered by high computational costs and latency, while resource-efficient Small Language Models (SLMs) typically lack the necessary reasoning capacity. Existing collaborative…

Computation and Language · Computer Science 2026-01-09 Chengsong Huang , Tong Zheng , Langlin Huang , Jinyuan Li , Haolin Liu , Jiaxin Huang

Diffusion Large Language Models (DLLMs) enable fully parallel token decoding but often remain impractical at inference time due to the many denoising iterations required to refine an information-free, fully masked initialization into…

Computation and Language · Computer Science 2025-12-23 Tongyuan Miao , Gary Huang , Kai Jun Han , Annie Jiang

Recent advances in masked diffusion language models (MDLMs) narrow the quality gap to autoregressive LMs, but their sampling remains expensive because generation requires many full-sequence denoising passes with a large Transformer and,…

Machine Learning · Computer Science 2026-04-14 Ivan Sedykh , Nikita Sorokin , Valentin Malykh

Diffusion language models promise parallel generation, yet still lag behind autoregressive (AR) models in quality. We stem this gap to a failure of introspective consistency: AR models agree with their own generations, while DLMs often do…

We present DMax, a new paradigm for efficient diffusion language models (dLLMs). It mitigates error accumulation in parallel decoding, enabling aggressive decoding parallelism while preserving generation quality. Unlike conventional masked…

Machine Learning · Computer Science 2026-05-18 Zigeng Chen , Gongfan Fang , Xinyin Ma , Ruonan Yu , Xinchao Wang

Federated inference enhances LLM performance in edge computing through weighted averaging of distributed model predictions. However, autoregressive LLM inference requires frequent full-model forward passes across workers, severely limiting…

Signal Processing · Electrical Eng. & Systems 2026-04-29 Ce Zheng , Xinghan Wang , Jiahong Ning , Yuxuan Shi , Ning Huang , Tingting Yang

In recent years, large language models (LLMs) have witnessed remarkable advancements, with the test-time scaling law consistently enhancing the reasoning capabilities. Through systematic evaluation and exploration of a diverse spectrum of…

Computation and Language · Computer Science 2025-11-03 Chenyang Shao , Sijian Ren , Fengli Xu , Yong Li

Diffusion language models (DLMs) have strong theoretical efficiency but are limited by fixed-length decoding and incompatibility with key-value (KV) caches. Block diffusion mitigates these issues, yet still enforces a fixed block size and…

Computation and Language · Computer Science 2025-09-30 Yangzhou Liu , Yue Cao , Hao Li , Gen Luo , Zhe Chen , Weiyun Wang , Xiaobo Liang , Biqing Qi , Lijun Wu , Changyao Tian , Yanting Zhang , Yuqiang Li , Tong Lu , Yu Qiao , Jifeng Dai , Wenhai Wang

Since local LLM inference on resource-constrained edge devices imposes a severe performance bottleneck, this paper proposes distributed prompt caching to enhance inference performance by cooperatively sharing intermediate processing states…

Machine Learning · Computer Science 2026-04-13 Hiroki Matsutani , Naoki Matsuda , Naoto Sugiura

Parallel decoding for diffusion LLMs (dLLMs) is difficult because each denoising step provides only token-wise marginal distributions, while unmasking multiple tokens simultaneously requires accounting for inter-token dependencies. We…

Machine Learning · Computer Science 2026-03-16 Bumjun Kim , Dongjae Jeon , Moongyu Jeon , Albert No

Large language models (LLMs) have achieved notable progress. Despite their success, next-token prediction (NTP), the dominant method for LLM training and inference, is constrained in both contextual coverage and inference efficiency due to…

Computation and Language · Computer Science 2025-09-23 Xiaohao Liu , Xiaobo Xia , Weixiang Zhao , Manyi Zhang , Xianzhi Yu , Xiu Su , Shuo Yang , See-Kiong Ng , Tat-Seng Chua

Speculative decoding has emerged as a widely adopted method to accelerate large language model inference without sacrificing the quality of the model outputs. While this technique has facilitated notable speed improvements by enabling…

Computation and Language · Computer Science 2025-02-12 Jacob K Christopher , Brian R Bartoldson , Tal Ben-Nun , Michael Cardei , Bhavya Kailkhura , Ferdinando Fioretto