English
Related papers

Related papers: Setting the Record Straight on Transformer Oversmo…

200 papers

Transformers serve as the foundational architecture for large language and video generation models, such as GPT, BERT, SORA and their successors. Empirical studies have demonstrated that real-world data and learning tasks exhibit…

Machine Learning · Computer Science 2026-05-19 Zhaiming Shen , Alex Havrilla , Rongjie Lai , Alexander Cloninger , Wenjing Liao

Data augmentation in feature space is effective to increase data diversity. Previous methods assume that different classes have the same covariance in their feature distributions. Thus, feature transform between different classes is…

Computer Vision and Pattern Recognition · Computer Science 2020-08-05 Yuke Zhu , Yan Bai , Yichen Wei

Structured data, which constitutes a significant portion of existing data types, has been a long-standing research topic in the field of machine learning. Various representation learning methods for tabular data have been proposed, ranging…

Artificial Intelligence · Computer Science 2023-12-19 Jayoung Kim , Yehjin Shin , Jeongwhan Choi , Hyowon Wi , Noseong Park

Overparameterization, the condition where models have more parameters than necessary to fit their training loss, is a crucial factor for the success of deep learning. However, the characteristics of the features learned by overparameterized…

Machine Learning · Computer Science 2024-07-02 Ahmet Cagri Duzgun , Samy Jelassi , Yuanzhi Li

Despite their central role in the success of foundational models and large-scale language modeling, the theoretical foundations governing the operation of Transformers remain only partially understood. Contemporary research has largely…

Machine Learning · Computer Science 2025-06-02 Sagar Ghosh , Kushal Bose , Swagatam Das

This paper reveals a novel linear characteristic exclusive to transformer decoders, including models such as GPT, LLaMA, OPT, BLOOM and others. We analyze embedding transformations between sequential layers, uncovering a near-perfect linear…

Transformers, renowned for their self-attention mechanism, have achieved state-of-the-art performance across various tasks in natural language processing, computer vision, time-series modeling, etc. However, one of the challenges with deep…

Machine Learning · Computer Science 2024-11-04 Jeongwhan Choi , Hyowon Wi , Jayoung Kim , Yehjin Shin , Kookjin Lee , Nathaniel Trask , Noseong Park

Transformers have the capacity to act as supervised learning algorithms: by properly encoding a set of labeled training ("in-context") examples and an unlabeled test example into an input sequence of vectors of the same dimension, the…

Machine Learning · Computer Science 2024-12-16 Spencer Frei , Gal Vardi

Understanding why trained Transformers generalize well is a fundamental problem in modern machine learning theory, and complexity-based generalization bounds provide a principled way to study this question. While existing norm-based bounds…

Machine Learning · Statistics 2026-05-11 Mana Sakai , Masaaki Imaizumi

Normalization and scaling are fundamental preprocessing steps in time series modeling, yet their role in Transformer-based models remains underexplored from a theoretical perspective. In this work, we present the first formal analysis of…

Machine Learning · Computer Science 2026-02-20 Sofiane Ennadir , Tianze Wang , Oleg Smirnov , Sahar Asadi , Lele Cao

We investigate how embedding dimension affects the emergence of an internal "world model" in a transformer trained with reinforcement learning to perform bubble-sort-style adjacent swaps. Models achieve high accuracy even with very small…

Machine Learning · Computer Science 2025-10-22 Brady Bhalla , Honglu Fan , Nancy Chen , Tony Yue YU

Neural autoregressive sequence models smear the probability among many possible sequences including degenerate ones, such as empty or repetitive sequences. In this work, we tackle one specific case where the model assigns a high probability…

Machine Learning · Computer Science 2021-12-23 Ilia Kulikov , Maksim Eremeev , Kyunghyun Cho

Prior work has explored directly regularizing the output distributions of probabilistic models to alleviate peaky (i.e. over-confident) predictions, a common sign of overfitting. This class of techniques, of which label smoothing is one,…

Computation and Language · Computer Science 2020-05-13 Clara Meister , Elizabeth Salesky , Ryan Cotterell

Perplexity -- a function measuring a model's overall level of "surprise" when encountering a particular output -- has gained significant traction in recent years, both as a loss function and as a simple-to-compute metric of model quality.…

Machine Learning · Computer Science 2026-02-02 Petar Veličković , Federico Barbero , Christos Perivolaropoulos , Simon Osindero , Razvan Pascanu

We conduct a systematic study of the approximation properties of Transformer for sequence modeling with long, sparse and complicated memory. We investigate the mechanisms through which different components of Transformer, such as the…

Machine Learning · Computer Science 2024-10-31 Mingze Wang , Weinan E

We develop an algorithm which, given a trained transformer model $\mathcal{M}$ as input, as well as a string of tokens $s$ of length $n_{fix}$ and an integer $n_{free}$, can generate a mathematical proof that $\mathcal{M}$ is…

Machine Learning · Computer Science 2025-05-27 Lev Stambler , Seyed Sajjad Nezhadi , Matthew Coudron

Transformer-based large language models have remarkable potential to accelerate design optimization for applications such as drug development and materials discovery. Self-supervised pretraining of transformer models requires large-scale…

Machine Learning · Computer Science 2023-10-27 Pei Zhang , Logan Kearney , Debsindhu Bhowmik , Zachary Fox , Amit K. Naskar , John Gounley

Graph convolutions have gained popularity due to their ability to efficiently operate on data with an irregular geometric structure. However, graph convolutions cause over-smoothing, which refers to representations becoming more similar…

Machine Learning · Computer Science 2024-09-04 Andreas Roth

In the software development process, model transformation is increasingly assimilated. However, systems being developed with model transformation sometimes grow in size and become complex. Meanwhile, the performance of model transformation…

Software Engineering · Computer Science 2020-04-21 Vijayshree Vijayshree , Markus Frank , Steffen Becker

Transformer-based models have delivered impressive results on many tasks, particularly vision and language tasks. In many model training situations, conventional configurations are typically adopted. For example, we often set the base model…

Machine Learning · Computer Science 2023-05-19 Fuzhao Xue , Jianghai Chen , Aixin Sun , Xiaozhe Ren , Zangwei Zheng , Xiaoxin He , Yongming Chen , Xin Jiang , Yang You