English
Related papers

Related papers: Dynamic Transformers Provide a False Sense of Effi…

200 papers

Large language models have achieved remarkable capabilities, but their practical deployment is hindered by significant computational costs. While adaptive computation methods like early-exiting promise to reduce these costs, they introduce…

Computation and Language · Computer Science 2025-12-16 Sangmin Bae

Recent advances in Transformer-based large language models (LLMs) have led to significant performance improvements across many tasks. These gains come with a drastic increase in the models' size, potentially leading to slow and costly use…

Computation and Language · Computer Science 2022-10-26 Tal Schuster , Adam Fisch , Jai Gupta , Mostafa Dehghani , Dara Bahri , Vinh Q. Tran , Yi Tay , Donald Metzler

Transformer-based pre-trained language models (PLMs) mostly suffer from excessive overhead despite their advanced capacity. For resource-constrained devices, there is an urgent need for a spatially and temporally efficient model which…

Computation and Language · Computer Science 2022-10-28 Bowen Shen , Zheng Lin , Yuanxin Liu , Zhengxiao Liu , Lei Wang , Weiping Wang

In this paper, we propose Patience-based Early Exit, a straightforward yet effective inference method that can be used as a plug-and-play technique to simultaneously improve the efficiency and robustness of a pretrained language model…

Computation and Language · Computer Science 2020-10-23 Wangchunshu Zhou , Canwen Xu , Tao Ge , Julian McAuley , Ke Xu , Furu Wei

Building efficient inference framework has gained increasing interests for research community. Early-exit models, a variant of LLMs, improves the inference efficiency of LLMs by skipping rest layers and directly generate output tokens when…

Computation and Language · Computer Science 2024-07-31 Ruijie Miao , Yihan Yan , Xinshuo Yao , Tong Yang

In this paper, we systematically evaluate the robustness of multi-exit language models against adversarial slowdown. To audit their robustness, we design a slowdown attack that generates natural adversarial text bypassing early-exit points.…

Machine Learning · Computer Science 2023-11-01 Zachary Coalson , Gabriel Ritter , Rakesh Bobba , Sanghyun Hong

Large-scale Transformer models bring significant improvements for various downstream vision language tasks with a unified architecture. The performance improvements come with increasing model size, resulting in slow inference speed and…

Computer Vision and Pattern Recognition · Computer Science 2023-04-04 Shengkun Tang , Yaqing Wang , Zhenglun Kong , Tianchi Zhang , Yao Li , Caiwen Ding , Yanzhi Wang , Yi Liang , Dongkuan Xu

In Large Language Model (LLM) inference, early-exit refers to stopping computation at an intermediate layer once the prediction is sufficiently confident, thereby reducing latency and cost. However, recent LLMs adopt improved pretraining…

Computation and Language · Computer Science 2026-03-26 Rui Wei , Rui Du , Hanfei Yu , Devesh Tiwari , Jian Li , Zhaozhuo Xu , Hao Wang

To tackle the high inference latency exhibited by autoregressive language models, previous studies have proposed an early-exiting framework that allocates adaptive computation paths for each token based on the complexity of generating the…

Computation and Language · Computer Science 2023-10-10 Sangmin Bae , Jongwoo Ko , Hwanjun Song , Se-Young Yun

The ability to dynamically adjust the computational load of neural models during inference in a resource aware manner is crucial for on-device processing scenarios, characterised by limited and time-varying computational resources.…

Computation and Language · Computer Science 2025-06-24 Maxence Lasbordes , Daniele Falavigna , Alessio Brutti

Early exiting is an effective paradigm for improving the inference efficiency of deep networks. By constructing classifiers with varying resource demands (the exits), such networks allow easy samples to be output at early exits, removing…

Computer Vision and Pattern Recognition · Computer Science 2022-09-20 Yizeng Han , Yifan Pu , Zihang Lai , Chaofei Wang , Shiji Song , Junfen Cao , Wenhui Huang , Chao Deng , Gao Huang

Adaptive inference is a simple method for reducing inference costs. The method works by maintaining multiple classifiers of different capacities, and allocating resources to each test instance according to its difficulty. In this work, we…

Computation and Language · Computer Science 2023-06-06 Daniel Rotem , Michael Hassid , Jonathan Mamou , Roy Schwartz

Agents powered by large language models (LLMs) have demonstrated strong planning and decision-making capabilities in complex embodied environments. However, such agents often suffer from inefficiencies in multi-turn interactions, frequently…

Computation and Language · Computer Science 2025-09-23 Qingyu Lu , Liang Ding , Siyi Cao , Xuebo Liu , Kanjian Zhang , Jinxia Zhang , Dacheng Tao

Deploying large language model inference remains challenging due to their high computational overhead. Early exit optimizes model inference by adaptively reducing the number of inference layers. Current methods typically train internal…

Computation and Language · Computer Science 2026-03-05 Lianming Huang , Shangyu Wu , Yufei Cui , Ying Xiong , Haibo Hu , Xue Liu , Tei-Wei Kuo , Nan Guan , Chun Jason Xue

Adaptive gradient methods, such as Adam and LAMB, have demonstrated excellent performance in the training of large language models. Nevertheless, the need for adaptivity requires maintaining second-moment estimates of the per-parameter…

Computation and Language · Computer Science 2023-08-08 Yang Luo , Xiaozhe Ren , Zangwei Zheng , Zhuo Jiang , Xin Jiang , Yang You

Machine learning (ML) inference platforms are tasked with balancing two competing goals: ensuring high throughput given many requests, and delivering low-latency responses to support interactive applications. Unfortunately, existing…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-09-25 Yinwei Dai , Rui Pan , Anand Iyer , Kai Li , Ravi Netravali

Early Exiting is one of the most popular methods to achieve efficient inference. Current early exiting methods adopt the (weighted) sum of the cross entropy loss of all internal classifiers during training, imposing all these classifiers to…

Computation and Language · Computer Science 2024-04-09 Ziqian Zeng , Yihuai Hong , Hongliang Dai , Huiping Zhuang , Cen Chen

DNNs are becoming less and less over-parametrised due to recent advances in efficient model design, through careful hand-crafted or NAS-based methods. Relying on the fact that not all inputs require the same amount of computation to yield a…

Machine Learning · Computer Science 2021-06-10 Stefanos Laskaridis , Alexandros Kouris , Nicholas D. Lane

Deep neural networks (DNN) are increasingly being accelerated on application-specific hardware such as the Google TPU designed especially for deep learning. Timing speculation is a promising approach to further increase the energy…

Machine Learning · Computer Science 2018-07-03 Jeff Zhang , Siddharth Garg

Early exiting is an effective paradigm for improving the inference efficiency of pre-trained language models (PLMs) by dynamically adjusting the number of executed layers for each sample. However, in most existing works, easy and hard…

Machine Learning · Computer Science 2024-12-19 Jianing He , Qi Zhang , Hongyun Zhang , Xuanjing Huang , Usman Naseem , Duoqian Miao
‹ Prev 1 2 3 10 Next ›