English
Related papers

Related papers: GradES: Significantly Faster Training in Transform…

200 papers

With increasing scale in model and dataset size, the training of deep neural networks becomes a massive computational burden. One approach to speed up the training process is Selective Backprop. For this approach, we perform a forward pass…

Machine Learning · Computer Science 2023-12-11 Lukas Balles , Cedric Archambeau , Giovanni Zappella

Deep neural networks have seen great success in recent years; however, training a deep model is often challenging as its performance heavily depends on the hyper-parameters used. In addition, finding the optimal hyper-parameter…

Attention sinks and massive activations are recurring and closely related phenomena in Transformer models. Existing explanations have largely focused on the forward pass, yet in pre-norm Transformers, large residual-stream norms play only…

Machine Learning · Computer Science 2026-05-07 Yihong Chen , Zhouchen Lin , Quanming Yao

Grokking is the phenomenon whereby, unlike the training performance, which peaks early in the training process, the test/generalization performance of a model stagnates over arbitrarily many epochs and then suddenly jumps to usually close…

Machine Learning · Computer Science 2026-05-19 Ali Saheb Pasand , Elvis Dohmatob

We develop an approach to efficiently grow neural networks, within which parameterization and optimization strategies are designed by considering their effects on the training dynamics. Unlike existing growing methods, which follow simple…

Machine Learning · Computer Science 2023-06-23 Xin Yuan , Pedro Savarese , Michael Maire

Pre-trained language models, such as BERT, have achieved significant accuracy gain in many natural language processing tasks. Despite its effectiveness, the huge number of parameters makes training a BERT model computationally very…

Computation and Language · Computer Science 2020-11-30 Cheng Yang , Shengnan Wang , Chao Yang , Yuechuan Li , Ru He , Jingqiao Zhang

Neural networks are trained primarily based on their inputs and outputs, without regard for their internal mechanisms. These neglected mechanisms determine properties that are critical for safety, like (i) transparency; (ii) the absence of…

Machine Learning · Computer Science 2024-12-02 Alex Cloud , Jacob Goldman-Wetzler , Evžen Wybitul , Joseph Miller , Alexander Matt Turner

The evolving sophistication and intricacies of Large Language Models (LLMs) yield unprecedented advancements, yet they simultaneously demand considerable computational resources and incur significant costs. To alleviate these challenges,…

Computation and Language · Computer Science 2023-10-03 Hongye Jin , Xiaotian Han , Jingfeng Yang , Zhimeng Jiang , Chia-Yuan Chang , Xia Hu

Feed-forward 3D Gaussian Splatting models offer fast single-pass reconstruction,but scaling them to match per-scene optimization quality is fundamentally hindered by the scarcity of large-scale 3D annotations. A practical compromise is…

Computer Vision and Pattern Recognition · Computer Science 2026-05-25 Yuke Li , Weihang Liu , Cheng Zhang , Yuefeng Zhang , Jiadi Cui , Zixuan Wang , Junran Ding , Haoyu Wu , Yujiao Shi , Jingyi Yu , Xin Lou

With the rapid growth in model size, fine-tuning the large pre-trained language model has become increasingly difficult due to its extensive memory usage. Previous works usually focus on reducing the number of trainable parameters in the…

Lateral inhibitory connections have been observed in the cortex of the biological brain, and has been extensively studied in terms of its role in cognitive functions. However, in the vanilla version of backpropagation in deep learning, all…

Computer Vision and Pattern Recognition · Computer Science 2022-08-16 Lei Jiang , Yongqing Liu , Shihai Xiao , Yansong Chua

Loss of trainability refers to a phenomenon in continual learning where parameter updates no longer make progress on the optimization objective, so accuracy stalls or degrades as the learning problem changes over time. In this paper, we…

Machine Learning · Computer Science 2025-12-11 Gunbir Singh Baveja , Alex Lewandowski , Mark Schmidt

The fully-convolutional siamese network based on template matching has shown great potentials in visual tracking. During testing, the template is fixed with the initial target feature and the performance totally relies on the general…

Computer Vision and Pattern Recognition · Computer Science 2019-09-17 Peixia Li , Boyu Chen , Wanli Ouyang , Dong Wang , Xiaoyun Yang , Huchuan Lu

Transformer-based neural models are used in many AI applications. Training these models is expensive, as it takes huge GPU resources and long duration. It is challenging because typical data like sentences have variable lengths, and…

Computation and Language · Computer Science 2022-06-17 Xiaohui Wang , Yang Wei , Ying Xiong , Guyue Huang , Xian Qian , Yufei Ding , Mingxuan Wang , Lei Li

Depth-adaptive neural networks can dynamically adjust depths according to the hardness of input words, and thus improve efficiency. The main challenge is how to measure such hardness and decide the required depths (i.e., layers) to conduct.…

Computation and Language · Computer Science 2020-12-17 Yijin Liu , Fandong Meng , Jie Zhou , Yufeng Chen , Jinan Xu

The success of gradient-based meta-learning is primarily attributed to its ability to leverage related tasks to learn task-invariant information. However, the absence of interactions between different tasks in the inner loop leads to…

Machine Learning · Computer Science 2023-12-15 Oscar Chang , Hod Lipson

During pretraining, the Pre-LayerNorm transformer suffers from a gradient magnitude mismatch: gradients at early layers are much larger than at later layers. These issues can be alleviated by our proposed NormFormer architecture, which adds…

Computation and Language · Computer Science 2021-11-02 Sam Shleifer , Jason Weston , Myle Ott

Large-scale pre-trained transformers have demonstrated remarkable success in various computer vision tasks. However, it is still highly challenging to fully fine-tune these models for downstream tasks due to their high computational and…

Computer Vision and Pattern Recognition · Computer Science 2023-04-27 Zhao Song , Ke Yang , Naiyang Guan , Junjie Zhu , Peng Qiao , Qingyong Hu

Safety post-training can improve the harmfulness and policy compliance of Large Language Models (LLMs), but it may also reduce general utility, a phenomenon often described as the \emph{alignment tax}. We study this trade-off through the…

Machine Learning · Computer Science 2026-05-13 Guanglong Sun , Siyuan Zhang , Liyuan Wang , Jun Zhu , Hang Su , Yi Zhong

Deep learning networks are typically trained by Stochastic Gradient Descent (SGD) methods that iteratively improve the model parameters by estimating a gradient on a very small fraction of the training data. A major roadblock faced when…

Machine Learning · Computer Science 2020-06-11 Tao Lin , Lingjing Kong , Sebastian U. Stich , Martin Jaggi