English
Related papers

Related papers: Representation Collapse in Machine Translation Thr…

200 papers

It is now a standard for neural network representations to be trained on large, publicly available datasets, and used for new problems. The reasons for why neural network representations have been so successful for transfer, however, are…

Machine Learning · Computer Science 2022-09-20 Ehsan Imani , Wei Hu , Martha White

Neural machine translation (NMT) models are typically trained with fixed-size input and output vocabularies, which creates an important bottleneck on their accuracy and generalization capability. As a solution, various studies proposed…

Computation and Language · Computer Science 2018-05-08 Duygu Ataman , Marcello Federico

We study the ability of foundation models to learn representations for classification that are transferable to new, unseen classes. Recent results in the literature show that representations learned by a single classifier over many classes…

Machine Learning · Computer Science 2022-01-05 Tomer Galanti , András György , Marcus Hutter

Vector quantization is a technique in machine learning that discretizes continuous representations into a set of discrete vectors. It is widely employed in tokenizing data representations for large language models, diffusion models, and…

Machine Learning · Computer Science 2026-03-19 Wenhao Zhao , Qiran Zou , Rushi Shah , Yudi Wu , Zhouhan Lin , Dianbo Liu

When training overparameterized deep networks for classification tasks, it has been widely observed that the learned features exhibit a so-called "neural collapse" phenomenon. More specifically, for the output features of the penultimate…

Machine Learning · Computer Science 2023-03-09 Can Yaras , Peng Wang , Zhihui Zhu , Laura Balzano , Qing Qu

In continual learning scenarios, catastrophic forgetting of previously learned tasks is a critical issue, making it essential to effectively measure such forgetting. Recently, there has been growing interest in focusing on representation…

Machine Learning · Computer Science 2025-06-13 Joonkyu Kim , Yejin Kim , Jy-yong Sohn

Online continual learning suffers from an underfitted solution due to insufficient training for prompt model update (e.g., single-epoch training). To address the challenge, we propose an efficient online continual learning method using the…

Computer Vision and Pattern Recognition · Computer Science 2024-04-03 Minhyuk Seo , Hyunseo Koh , Wonje Jeung , Minjae Lee , San Kim , Hankook Lee , Sungjun Cho , Sungik Choi , Hyunwoo Kim , Jonghyun Choi

Self-supervised representation learning is central to modern machine learning because it extracts structured latent features from unlabeled data and enables robust transfer across tasks and domains. However, it can suffer from…

Disordered Systems and Neural Networks · Physics 2026-04-14 Louie Hong Yao , Yuhao Li , Shengchao Liu

We seek to understand how the representations of individual tokens and the structure of the learned feature space evolve between layers in deep neural networks under different learning objectives. We focus on the Transformers for our…

Computation and Language · Computer Science 2019-09-05 Elena Voita , Rico Sennrich , Ivan Titov

Neural Collapse (NC) is a well-known phenomenon of deep neural networks in the terminal phase of training (TPT). It is characterized by the collapse of features and classifier into a symmetrical structure, known as simplex equiangular tight…

Machine Learning · Computer Science 2023-10-13 Peifeng Gao , Qianqian Xu , Yibo Yang , Peisong Wen , Huiyang Shao , Zhiyong Yang , Bernard Ghanem , Qingming Huang

Prototypical self-supervised learning methods consistently suffer from partial prototype collapse, where multiple prototypes converge to nearly identical representations. This undermines their central purpose -- providing diverse and…

Modern practice for training classification deepnets involves a Terminal Phase of Training (TPT), which begins at the epoch where training error first vanishes; During TPT, the training error stays effectively zero while training loss is…

Machine Learning · Computer Science 2020-09-23 Vardan Papyan , X. Y. Han , David L. Donoho

Neural Machine Translation (NMT) is resource intensive. We design a quantization procedure to compress NMT models better for devices with limited hardware capability. Because most neural network parameters are near zero, we employ…

Computation and Language · Computer Science 2019-09-23 Alham Fikri Aji , Kenneth Heafield

Continual Learning research typically focuses on tackling the phenomenon of catastrophic forgetting in neural networks. Catastrophic forgetting is associated with an abrupt loss of knowledge previously learned by a model when the task, or…

Machine Learning · Computer Science 2022-04-06 MohammadReza Davari , Nader Asadi , Sudhir Mudur , Rahaf Aljundi , Eugene Belilovsky

Augmenting neural machine translation with external memory at decoding time, in the form of k-nearest neighbors machine translation ($k$-NN MT), is a well-established strategy for increasing translation performance. $k$-NN MT retrieves a…

Computation and Language · Computer Science 2025-09-23 Evgeniia Tokarchuk , Sergey Troshin , Vlad Niculae

In contrastive representation learning, data representation is trained so that it can classify the image instances even when the images are altered by augmentations. However, depending on the datasets, some augmentations can damage the…

Machine Learning · Statistics 2021-11-16 Masanori Koyama , Kentaro Minami , Takeru Miyato , Yarin Gal

Next-token prediction (NTP) over large text corpora has become the go-to paradigm to train large language models. Yet, it remains unclear how NTP influences the mapping of linguistic patterns to geometric properties of the resulting model…

Computation and Language · Computer Science 2025-02-20 Yize Zhao , Tina Behnia , Vala Vakilian , Christos Thrampoulidis

Training deep neural networks for classification often includes minimizing the training loss beyond the zero training error point. In this phase of training, a "neural collapse" behavior has been observed: the variability of features…

Machine Learning · Computer Science 2023-05-30 Tom Tirer , Haoxiang Huang , Jonathan Niles-Weed

We argue that translation quality alone is not a sufficient metric for measuring knowledge transfer in multilingual neural machine translation. To support this claim, we introduce Representational Transfer Potential (RTP), which measures…

Computation and Language · Computer Science 2023-12-05 David Stap , Vlad Niculae , Christof Monz

Deep neural networks that achieve remarkable performance in image classification have previously been shown to be easily fooled by tiny transformations such as a one pixel translation of the input image. In order to address this problem,…

Computer Vision and Pattern Recognition · Computer Science 2024-04-11 Ofir Shifman , Yair Weiss