English
Related papers

Related papers: Sink-Aware Pruning for Diffusion Language Models

200 papers

Automatic Speech Recognition (ASR) transcripts often contain disfluencies, such as fillers, repetitions, and false starts, which reduce readability and hinder downstream applications like chatbots and voice assistants. If left unaddressed,…

Computation and Language · Computer Science 2026-05-13 Deepak Kumar , Baban Gain , Asif Ekbal

How can we accelerate large language models(LLMs) without sacrificing accuracy? The slow inference speed of LLMs hinders us to benefit from their remarkable performance in diverse applications. This is mainly because numerous sublayers are…

Computation and Language · Computer Science 2025-06-05 Seungcheol Park , Sojin Lee , Jongjin Kim , Jinsik Lee , Hyunjik Jo , U Kang

Deep learning drives a new wave in computing systems and triggers the automation of increasingly complex problems. In particular, Large Language Models (LLMs) have significantly advanced cognitive tasks, often matching or even surpassing…

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…

Computation and Language · Computer Science 2025-12-08 Tianyi Li , Mingda Chen , Bowei Guo , Zhiqiang Shen

Deploying Large Language Models (LLMs) in streaming applications such as multi-round dialogue, where long interactions are expected, is urgently needed but poses two major challenges. Firstly, during the decoding stage, caching previous…

Computation and Language · Computer Science 2024-04-09 Guangxuan Xiao , Yuandong Tian , Beidi Chen , Song Han , Mike Lewis

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…

This paper explores pruning attention heads as a post-processing bias mitigation method for large language models (LLMs). Modern AI systems such as LLMs are expanding into sensitive social contexts where fairness concerns become especially…

Artificial Intelligence · Computer Science 2025-11-26 Vishnu Asutosh Dasu , Md Rafi ur Rashid , Vipul Gupta , Saeid Tizpaz-Niari , Gang Tan

Pruning provides a practical solution to reduce the resources required to run large language models (LLMs) to benefit from their effective capabilities as well as control their cost for training and inference. Research on LLM pruning often…

Computation and Language · Computer Science 2025-10-28 Yuanhe Tian , Junjie Liu , Xican Yang , Haishan Ye , Yan Song

Large Language Models (LLMs) exhibit substantial parameter redundancy, particularly in Feed-Forward Networks (FFNs). Existing pruning methods suffer from two primary limitations. First, reliance on dataset-specific calibration introduces…

Computation and Language · Computer Science 2026-02-02 Abhishek Tyagi , Yunuo Cen , Shrey Dhorajiya , Bharadwaj Veeravalli , Xuanyao Fong

This paper introduces LLM-Streamline, a pioneer work on layer pruning for large language models (LLMs). It is based on the observation that different layers have varying impacts on hidden states, enabling the identification of less…

Computation and Language · Computer Science 2025-01-28 Xiaodong Chen , Yuxuan Hu , Jing Zhang , Yanling Wang , Cuiping Li , Hong Chen

Diffusion language models (DLMs) have recently demonstrated capabilities that complement standard autoregressive (AR) models, particularly in non-sequential generation and bidirectional editing. Although recent work has shown that…

Machine Learning · Computer Science 2026-05-11 Fred Zhangzhi Peng , Alexis Fox , Anru R. Zhang , Alexander Tong

Due to the large number of parameters, the inference phase of Large Language Models (LLMs) is resource-intensive. Unlike traditional model compression, which needs retraining, recent dynamic computation methods show that not all components…

Computation and Language · Computer Science 2025-11-27 Siqi Fan , Xuezhi Fang , Xingrun Xing , Peng Han , Shuo Shang , Yequan Wang

As the computational needs of Large Vision-Language Models (LVLMs) increase, visual token pruning has proven effective in improving inference speed and memory efficiency. Traditional pruning methods in LVLMs predominantly focus on attention…

Computer Vision and Pattern Recognition · Computer Science 2025-03-12 Bozhi Luan , Wengang Zhou , Hao Feng , Zhe Wang , Xiaosong Li , Houqiang Li

Pre-trained language models achieve superior performance but are computationally expensive. Techniques such as pruning and knowledge distillation have been developed to reduce their sizes and latencies. In this work, we propose a structured…

Computation and Language · Computer Science 2023-05-19 Ziqing Yang , Yiming Cui , Xin Yao , Shijin Wang

Large language models (LLMs) face significant deployment challenges due to their massive computational demands. % While pruning offers a promising compression solution, existing methods suffer from two critical limitations: (1) They neglect…

Machine Learning · Computer Science 2026-04-01 Lang Xiong , Ning Liu , Ao Ren , Yuheng Bai , Haining Fang , BinYan Zhang , Zhe Jiang , Yujuan Tan , Duo Liu

In spite of strong performance achieved by LLMs, the costs of their deployment are unaffordable. For the compression of LLMs, gradient-based pruning methods present promising effectiveness. However, in these methods, the gradient…

Computation and Language · Computer Science 2025-06-16 Hourun Zhu , Chengchao Shen

Long-context inference for Large Language Models (LLMs) is heavily limited by high computational demands. While several existing methods optimize attention computation, they still process the full set of hidden states at each layer,…

Computation and Language · Computer Science 2025-11-25 Lingkun Long , Rubing Yang , Yushi Huang , Desheng Hui , Ao Zhou , Jianlei Yang

Diffusion Large Language Models (dLLMs) offer a compelling paradigm for natural language generation, leveraging parallel decoding and bidirectional attention to achieve superior global coherence compared to autoregressive models. While…

Machine Learning · Computer Science 2026-01-28 Zhongyu Xiao , Zhiwei Hao , Jianyuan Guo , Yong Luo , Jia Liu , Jie Xu , Han Hu

Large multimodal models (LMMs) "see" images by leveraging the attention mechanism between text and visual tokens in the transformer decoder. Ideally, these models should focus on key visual information relevant to the text token. However,…

Computer Vision and Pattern Recognition · Computer Science 2025-03-06 Seil Kang , Jinyeong Kim , Junhyeok Kim , Seong Jae Hwang

Large language models (LLMs) achieve remarkable performance across tasks but incur substantial computational costs due to their deep, multi-layered architectures. Layer pruning has emerged as a strategy to alleviate these inefficiencies,…

Computation and Language · Computer Science 2025-06-05 Anhao Zhao , Fanghua Ye , Yingqi Fan , Junlong Tong , Zhiwei Fei , Hui Su , Xiaoyu Shen