English
Related papers

Related papers: Training Dynamics of Transformers to Recognize Wor…

200 papers

We show that a neural network originally designed for language processing can learn the dynamical rules of a stochastic system by observation of a single dynamical trajectory of the system, and can accurately predict its emergent behavior…

Statistical Mechanics · Physics 2022-02-18 Corneel Casert , Isaac Tamblyn , Stephen Whitelam

Linear layers in neural networks (NNs) trained by gradient descent can be expressed as a key-value memory system which stores all training datapoints and the initial weights, and produces outputs using unnormalised dot attention over the…

Machine Learning · Computer Science 2022-06-20 Kazuki Irie , Róbert Csordás , Jürgen Schmidhuber

Large language models (LLMs) based on transformer architectures are typically described through collections of architectural components and training procedures, obscuring their underlying computational structure. This review article…

Machine Learning · Computer Science 2026-02-03 Vikram Krishnamurthy

Although transformers have demonstrated impressive capabilities for in-context learning (ICL) in practice, theoretical understanding of the underlying mechanism that allows transformers to perform ICL is still in its infancy. This work aims…

Machine Learning · Computer Science 2025-05-30 Wei Shen , Ruida Zhou , Jing Yang , Cong Shen

When optimizing over-parameterized models, such as deep neural networks, a large set of parameters can achieve zero training error. In such cases, the choice of the optimization algorithm and its respective hyper-parameters introduces…

Machine Learning · Computer Science 2019-12-06 Gauthier Gidel , Francis Bach , Simon Lacoste-Julien

Transformer is a ubiquitous model for natural language processing and has attracted wide attentions in computer vision. The attention maps are indispensable for a transformer model to encode the dependencies among input tokens. However,…

Machine Learning · Computer Science 2021-02-26 Yujing Wang , Yaming Yang , Jiangang Bai , Mingliang Zhang , Jing Bai , Jing Yu , Ce Zhang , Gao Huang , Yunhai Tong

Fitting neural networks often resorts to stochastic (or similar) gradient descent which is a noise-tolerant (and efficient) resolution of a gradient descent dynamics. It outputs a sequence of networks parameters, which sequence evolves…

Machine Learning · Statistics 2021-04-15 Gabriel Turinici

Neural networks trained via gradient descent with random initialization and without any regularization enjoy good generalization performance in practice despite being highly overparametrized. A promising direction to explain this phenomenon…

Machine Learning · Computer Science 2022-05-17 Hancheng Min , Salma Tarmoun , Rene Vidal , Enrique Mallada

Transformers can perform in-context classification from a few labeled examples, yet the inference-time algorithm remains opaque. We study multi-class linear classification in the hard no-margin regime and make the computation identifiable…

Machine Learning · Computer Science 2026-04-20 Patrick Lutz , Themistoklis Haris , Arjun Chandra , Aditya Gangrade , Venkatesh Saligrama

We study the implicit bias of gradient flow (i.e., gradient descent with infinitesimal step size) on linear neural network training. We propose a tensor formulation of neural networks that includes fully-connected, diagonal, and…

Machine Learning · Computer Science 2021-09-13 Chulhee Yun , Shankar Krishnan , Hossein Mobahi

Large transformer-based models are able to perform in-context few-shot learning, without being explicitly trained for it. This observation raises the question: what aspects of the training regime lead to this emergent behavior? Here, we…

Transformer-based language models are trained on large datasets to predict the next token given an input sequence. Despite this simple training objective, they have led to revolutionary advances in natural language processing. Underlying…

Machine Learning · Computer Science 2024-03-14 Yingcong Li , Yixiao Huang , M. Emrullah Ildiz , Ankit Singh Rawat , Samet Oymak

The dominating NLP paradigm of training a strong neural predictor to perform one task on a specific dataset has led to state-of-the-art performance in a variety of applications (eg. sentiment classification, span-prediction based question…

Computation and Language · Computer Science 2021-09-06 Paul Michel

Large language models (LLMs) often exhibit abrupt emergent behavior, whereby new abilities arise at certain points during their training. This phenomenon, commonly referred to as a ''phase transition'', remains poorly understood. In this…

Computation and Language · Computer Science 2025-04-01 Yuko Nakagi , Keigo Tada , Sota Yoshino , Shinji Nishimoto , Yu Takagi

We investigate the ability of transformers to perform in-context reinforcement learning (ICRL), where a model must infer and execute learning algorithms from trajectory data without parameter updates. We show that a linear self-attention…

Machine Learning · Statistics 2026-05-08 Haodong Liang , Lifeng Lai

Transformers are extremely successful machine learning models whose mathematical properties remain poorly understood. Here, we rigorously characterize the behavior of transformers with hardmax self-attention and normalization sublayers as…

Computation and Language · Computer Science 2026-05-14 Albert Alcalde , Giovanni Fantuzzi , Enrique Zuazua

Convolutional neural networks are widely used in imaging and image recognition. Learning such networks from training data leads to the minimization of a non-convex function. This makes the analysis of standard optimization methods such as…

Optimization and Control · Mathematics 2026-01-14 Jona-Maria Diederen , Holger Rauhut , Ulrich Terstiege

Sequence modelling requires determining which past tokens are causally relevant from the context and their importance: a process inherent to the attention layers in transformers, yet whose underlying learned mechanisms remain poorly…

Machine Learning · Computer Science 2026-04-14 Francesco D'Angelo , Nicolas Flammarion

Semantic associations such as the link between "bird" and "flew" are foundational for language modeling as they enable models to go beyond memorization and instead generalize and generate coherent text. Understanding how these associations…

Computation and Language · Computer Science 2026-05-14 Shawn Im , Changdae Oh , Zhen Fang , Sharon Li

Large language models are powerful but costly. We ask whether meta-learning can make the pretraining of small language models not only better but also more interpretable. We integrate first-order MAML with subset-masked LM pretraining,…

Computation and Language · Computer Science 2025-11-10 David Demitri Africa , Yuval Weiss , Paula Buttery , Richard Diehl Martinez