中文
相关论文

相关论文: Confidence-Based Decoding is Provably Efficient fo…

200 篇论文

Diffusion language models (DLMs) are promising alternatives to autoregressive language models (ARMs), yet the intrinsic differences in their generated text remain underexplored. We first find empirically that off-the-shelf DLMs exhibit…

计算与语言 · 计算机科学 2026-05-14 Zeyang Zhang , Chengwei Liang , Xingyan Chen , Meiqi Gu , Minrui Luo , Jingzhao Zhang , Tianxing He

Diffusion language models (DLMs) have emerged as a promising alternative to the long-dominant autoregressive (AR) paradigm, offering a parallelable decoding process that could yield greater efficiency. Yet, in practice, current open-source…

计算与语言 · 计算机科学 2025-11-11 Han Peng , Peiyu Liu , Zican Dong , Daixuan Cheng , Junyi Li , Yiru Tang , Shuo Wang , Wayne Xin Zhao

While LLM-based Automatic Speech Recognition (ASR) achieves high accuracy, its speed is limited by sequential autoregressive decoding. Diffusion Language Models (DLMs) offer a parallel alternative, yet their decoding strategies remain…

音频与语音处理 · 电气工程与系统科学 2026-05-29 Jeong Hun Yeo , Minsu Kim , Hyeongseop Rha , Yong Man Ro

Diffusion Language Models (DLMs) are rapidly emerging as a powerful and promising alternative to the dominant autoregressive (AR) paradigm. By generating tokens in parallel through an iterative denoising process, DLMs possess inherent…

计算与语言 · 计算机科学 2025-12-08 Tianyi Li , Mingda Chen , Bowei Guo , Zhiqiang Shen

Auto-regressive models (ARMs) have established a dominant paradigm in language modeling. However, their strictly sequential decoding paradigm imposes fundamental constraints on both inference efficiency and modeling flexibility. To address…

计算与语言 · 计算机科学 2026-04-13 Yuyan Zhou , Kai Syun Hou , Weiyu Chen , James Kwok

Diffusion-based language models (dLLMs) have emerged as a promising alternative to autoregressive language models, offering the potential for parallel token generation and bidirectional context modeling. However, harnessing this flexibility…

计算与语言 · 计算机科学 2026-05-28 Jiyeon Kim , Sungik Choi , Yongrae Jo , Moontae Lee , Minjoon Seo

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…

机器学习 · 计算机科学 2026-01-30 Shuibai Zhang , Fred Zhangzhi Peng , Yiheng Zhang , Jin Pan , Grigorios G. Chrysos

Diffusion Language Models (DLMs) generate text by iteratively denoising a masked sequence, repeatedly deciding which positions to commit at each step. Standard decoding follows a greedy rule: unmask the most confident positions, yet this…

计算与语言 · 计算机科学 2026-02-26 Mingyu Cao , Alvaro H. C. Correia , Christos Louizos , Shiwei Liu , Lu Yin

Diffusion Large Language Models (dLLMs) have emerged as a promising alternative to autoregressive (AR) LLMs for text generation, with the potential to decode multiple tokens in a single iteration. However, none of the existing open-source…

机器学习 · 计算机科学 2025-08-14 Xu Wang , Chenkai Xu , Yijie Jin , Jiachun Jin , Hao Zhang , Zhijie Deng

Diffusion large language models (dLLMs) theoretically permit token decoding in arbitrary order, a flexibility that could enable richer exploration of reasoning paths than autoregressive (AR) LLMs. In practice, however, random-order decoding…

计算与语言 · 计算机科学 2026-04-02 Liancheng Fang , Aiwei Liu , Henry Peng Zou , Yankai Chen , Enze Ma , Leyi Pan , Chunyu Miao , Wei-Chieh Huang , Xue Liu , Philip S. Yu

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…

计算与语言 · 计算机科学 2025-12-03 Kecheng Chen , Ziru Liu , Xijia Tao , Hui Liu , Xinyu Fu , Suiyun Zhang , Dandan Tu , Lingpeng Kong , Rui Liu , Haoliang Li

Diffusion language models (DLMs) have recently emerged as a strong alternative to autoregressive models by enabling parallel text generation. To improve inference efficiency and KV-cache compatibility, prior work commonly adopts block-based…

计算与语言 · 计算机科学 2026-01-21 Yingte Shu , Yuchuan Tian , Chao Xu , Yunhe Wang , Hanting Chen

Diffusion language models (DLMs) have recently emerged as a promising alternative to autoregressive (AR) approaches, enabling parallel token generation beyond a rigid left-to-right order. Despite growing empirical success, the theoretical…

机器学习 · 计算机科学 2026-02-24 Yunxiao Zhao , Changxiao Cai

Large Language Models (LLMs) have achieved state-of-the-art performance on a broad range of Natural Language Processing (NLP) tasks, including document processing and code generation. Autoregressive Language Models (ARMs), which generate…

Diffusion Language Models (DLMs) have recently emerged as a strong alternative to autoregressive language models (LMs). DLMs offer comparable accuracy with faster inference speed via parallel decoding. However, standard DLM decoding…

机器学习 · 计算机科学 2025-11-27 Hengyu Fu , Baihe Huang , Virginia Adams , Charles Wang , Venkat Srinivasan , Jiantao Jiao

Large Language Diffusion Models (LLDMs) benefit from a flexible decoding mechanism that enables parallelized inference and controllable generations over autoregressive models. Yet such flexibility introduces a critical challenge: inference…

机器学习 · 计算机科学 2025-12-05 Yichuan Mo , Quan Chen , Mingjie Li , Zeming Wei , Yisen Wang

Masked diffusion language models (MDLMs) have recently emerged as a new paradigm in language modeling, offering flexible generation dynamics and enabling efficient parallel decoding. However, existing decoding strategies for pre-trained…

计算与语言 · 计算机科学 2026-03-17 Xueyu Zhou , Yangrong Hu , Jian Huang

Diffusion large language models (dLLMs) have shown advantages in text generation, particularly due to their inherent ability for parallel decoding. However, constrained by the quality--speed trade-off, existing inference solutions adopt…

计算与语言 · 计算机科学 2026-02-09 Lizhuo Luo , Zhuoran Shi , Jiajun Luo , Zhi Wang , Shen Ren , Wenya Wang , Tianwei Zhang

Diffusion large language models (dLLMs) represent a significant advancement in text generation, offering parallel token decoding capabilities. However, existing open-source implementations suffer from quality-speed trade-offs that impede…

计算与语言 · 计算机科学 2025-10-09 Fanheng Kong , Jingyuan Zhang , Yahui Liu , Zirui Wu , Yu Tian , Victoria W. , Guorui Zhou

Augmenting Large Language Models (LLMs) with retrieved external knowledge has proven effective for improving the factual accuracy of generated responses. Despite their success, retrieval-augmented LLMs still face the distractibility issue,…

计算与语言 · 计算机科学 2025-02-18 Zexuan Qiu , Zijing Ou , Bin Wu , Jingjing Li , Aiwei Liu , Irwin King
‹ 上一页 1 2 3 10 下一页 ›