English
Related papers

Related papers: Post-LayerNorm Is Back: Stable, ExpressivE, and De…

200 papers

Residual neural networks (ResNets) are a promising class of deep neural networks that have shown excellent performance for a number of learning tasks, e.g., image classification and recognition. Mathematically, ResNet architectures can be…

Optimization and Control · Mathematics 2019-07-26 S. Günther , L. Ruthotto , J. B. Schroder , E. C. Cyr , N. R. Gauger

The swift advancement in the scales and capabilities of Large Language Models (LLMs) positions them as promising tools for a variety of downstream tasks. In addition to the pursuit of better performance and the avoidance of violent feedback…

Computation and Language · Computer Science 2023-09-28 Haoyu Wang , Guozheng Ma , Cong Yu , Ning Gui , Linrui Zhang , Zhiqi Huang , Suwei Ma , Yongzhe Chang , Sen Zhang , Li Shen , Xueqian Wang , Peilin Zhao , Dacheng Tao

Parameter-efficient transfer learning (PETL) has emerged as a flourishing research field for adapting large pre-trained models to downstream tasks, greatly reducing trainable parameters while grappling with memory challenges during…

Computer Vision and Pattern Recognition · Computer Science 2024-07-11 Haiwen Diao , Bo Wan , Xu Jia , Yunzhi Zhuge , Ying Zhang , Huchuan Lu , Long Chen

Fine-tuning pre-trained models has become the standard approach to endow them with specialized knowledge, but it poses fundamental challenges. In particular, \textit{(i)} fine-tuning often leads to catastrophic forgetting, where…

By replacing standard non-linearities with polynomial activations, Polynomial Neural Networks (PNNs) are pivotal for applications such as privacy-preserving inference via Homomorphic Encryption (HE). However, training PNNs effectively…

Machine Learning · Computer Science 2025-05-20 Forsad Al Hossain , Tauhidur Rahman

LLM post-training typically propagates task gradients through the full depth of the model. Although this end-to-end structure is simple and general, it couples task adaptation to full-depth activation storage, long-range backward…

Computation and Language · Computer Science 2026-05-11 Hengyu Shi , Tianyang Han , Peizhe Wang , Zhiling Wang , Xu Yang , Junhao Su

Large language models (LLMs) are routinely pre-trained on billions of tokens, only to start the process over again once new data becomes available. A much more efficient solution is to continually pre-train these models, saving significant…

Deep neural networks have a good success record and are thus viewed as the best architecture choice for complex applications. Their main shortcoming has been, for a long time, the vanishing gradient which prevented the numerical…

Machine Learning · Computer Science 2024-05-02 Bernhard Bermeitinger , Tomas Hrycej , Siegfried Handschuh

While Large Language Models (LLMs) demonstrate strong performance across domains, their long-context capabilities are limited by transient neural activations causing information decay and unstructured feed-forward network (FFN) weights…

Neurons and Cognition · Quantitative Biology 2026-04-13 Kangcong Li , Peng Ye , Chongjun Tu , Lin Zhang , Chunfeng Song , Jiamin Wu , Tao Yang , Qihao Zheng , Tao Chen

Transformers with linear recurrent modeling offer linear-time training and constant-memory inference. Despite their demonstrated efficiency and performance, pretraining such non-standard architectures from scratch remains costly and risky.…

Computation and Language · Computer Science 2025-05-08 Disen Lan , Weigao Sun , Jiaxi Hu , Jusen Du , Yu Cheng

Training large language models (LLMs) typically involves pre-training on massive corpora, only to restart the process entirely when new data becomes available. A more efficient and resource-conserving approach would be continual…

Deep neural network can easily overfit to even noisy labels due to its high capacity, which degrades the generalization performance of a model. To overcome this issue, we propose a new approach for learning from noisy labels (LNL) via…

Computer Vision and Pattern Recognition · Computer Science 2023-04-20 Seulki Park , Hwanjun Song , Daeho Um , Dae Ung Jo , Sangdoo Yun , Jin Young Choi

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

Recently, DeepNorm scales Transformers into extremely deep (i.e., 1000 layers) and reveals the promising potential of deep scaling. To stabilize the training of deep models, DeepNorm (Wang et al., 2022) attempts to constrain the model…

Machine Learning · Computer Science 2023-05-05 Yijin Liu , Xianfeng Zeng , Fandong Meng , Jie Zhou

Low-rank architectures have become increasingly important for efficient large language model (LLM) pre-training, providing substantial reductions in both parameter complexity and memory/computational demands. Despite these advantages,…

Machine Learning · Computer Science 2026-05-14 Boao Kong , Junzhu Liang , Yuxi Liu , Renjia Deng , Kun Yuan

The efficiency of large language models (LLMs) is fundamentally limited by their sequential, token-by-token generation process. We argue that overcoming this bottleneck requires a new design axis for LLM scaling: increasing the semantic…

Computation and Language · Computer Science 2025-11-03 Chenze Shao , Darren Li , Fandong Meng , Jie Zhou

The prevailing assumption of an exponential decay in large language model (LLM) reliability with sequence length, predicated on independent per-token error probabilities, posits an inherent limitation for long autoregressive outputs. Our…

Computation and Language · Computer Science 2026-05-07 Mikhail L. Arbuzov , Sisong Bei , Ziwei Dong , Dmitri Kalaev , Alexey A. Shvets

Recurrent Neural Networks (RNN), Long Short-Term Memory Networks (LSTM), and Memory Networks which contain memory are popularly used to learn patterns in sequential data. Sequential data has long sequences that hold relationships. RNN can…

Computation and Language · Computer Science 2019-04-22 Anupiya Nugaliyadde , Kok Wai Wong , Ferdous Sohel , Hong Xie

Depth is one of the key factors behind the success of convolutional neural networks (CNNs). Since ResNet, we are able to train very deep CNNs as the gradient vanishing issue has been largely addressed by the introduction of skip…

Computer Vision and Pattern Recognition · Computer Science 2020-02-18 Yong Guo , Jian Chen , Qing Du , Anton Van Den Hengel , Qinfeng Shi , Mingkui Tan

Deep residual architectures, such as ResNet and the Transformer, have enabled models of unprecedented depth, yet a formal understanding of why depth is so effective remains an open question. A popular intuition, following Veit et al.…

Machine Learning · Computer Science 2025-10-09 Benoit Dherin , Michael Munn