English
Related papers

Related papers: Training Dynamics of Transformers to Recognize Wor…

200 papers

Practitioners have consistently observed three puzzling phenomena in transformer-based large language models (LLMs): attention sinks, value-state drains, and residual-state peaks, collectively referred to as extreme-token phenomena. These…

Machine Learning · Computer Science 2024-11-08 Tianyu Guo , Druv Pai , Yu Bai , Jiantao Jiao , Michael I. Jordan , Song Mei

We propose a novel framework for analyzing the dynamics of distribution shift in real-world systems that captures the feedback loop between learning algorithms and the distributions on which they are deployed. Prior work largely models…

Machine Learning · Computer Science 2023-10-31 Lauren Conger , Franca Hoffmann , Eric Mazumdar , Lillian Ratliff

Continual learning, the ability of a model to adapt to an ongoing sequence of tasks without forgetting earlier ones, is a central goal of artificial intelligence. To better understand its underlying mechanisms, we study the limitations of…

Machine Learning · Statistics 2026-04-21 Hossein Taheri , Avishek Ghosh , Arya Mazumdar

Learning in Deep Neural Networks (DNN) takes place by minimizing a non-convex high-dimensional loss function, typically by a stochastic gradient descent (SGD) strategy. The learning process is observed to be able to find good minimizers…

Machine Learning · Computer Science 2020-03-12 Carlo Baldassi , Fabrizio Pittorino , Riccardo Zecchina

Diffusion and flow-based generative models have achieved remarkable success in domains such as image synthesis, video generation, and natural language modeling. In this work, we extend these advances to weight space learning by leveraging…

Machine Learning · Computer Science 2025-10-17 Daniel Saragih , Deyu Cao , Tejas Balaji

We present a theoretical and empirical study of the gradient dynamics of overparameterized shallow ReLU networks with one-dimensional input, solving least-squares interpolation. We show that the gradient dynamics of such networks are…

Machine Learning · Computer Science 2019-06-20 Francis Williams , Matthew Trager , Claudio Silva , Daniele Panozzo , Denis Zorin , Joan Bruna

In the Transformer model, "self-attention" combines information from attended embeddings into the representation of the focal embedding in the next layer. Thus, across layers of the Transformer, information originating from different tokens…

Machine Learning · Computer Science 2020-06-02 Samira Abnar , Willem Zuidema

We investigate the transformer's capability to simulate the training process of deep models via in-context learning (ICL), i.e., in-context deep learning. Our key contribution is providing a positive example of using a transformer to train…

Machine Learning · Computer Science 2025-04-15 Weimin Wu , Maojiang Su , Jerry Yao-Chieh Hu , Zhao Song , Han Liu

The transformer is the most popular neural architecture for language modeling. The cornerstone of the transformer is its global attention mechanism, which lets the model aggregate information from all preceding tokens before generating the…

Computation and Language · Computer Science 2026-05-20 Jiaoda Li , Ryan Cotterell

Multimodal learning has been lacking principled ways of combining information from different modalities and learning a low-dimensional manifold of meaningful representations. We study multimodal learning and sensor fusion from a latent…

Machine Learning · Computer Science 2019-04-24 Lijiang Guo

This paper introduces feature gradient flow, a new technique for interpreting deep learning models in terms of features that are understandable to humans. The gradient flow of a model locally defines nonlinear coordinates in the input data…

Image and Video Processing · Electrical Eng. & Systems 2023-07-26 Yinzhu Jin , Jonathan C. Garneau , P. Thomas Fletcher

Reinforcement learning (RL) has become a central post-training paradigm for large language models (LLMs), but its performance is highly sensitive to the quality of training problems. This sensitivity stems from the non-stationarity of RL:…

Machine Learning · Computer Science 2026-02-26 Ningyuan Yang , Weihua Du , Weiwei Sun , Sean Welleck , Yiming Yang

Phase transitions mark qualitative reorganizations of collective behavior, yet identifying their boundaries remains challenging whenever analytic solutions are absent and conventional simulations fail. Here we introduce learnability as a…

Materials Science · Physics 2025-10-10 Şener Özönder

Deep learning (DL) models based on the transformer architecture have revolutionized many DL applications such as large language models (LLMs), vision transformers, audio generation, and time series prediction. Much of this progress has been…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-08-20 Quentin Anthony , Benjamin Michalowicz , Jacob Hatef , Lang Xu , Mustafa Abduljabbar , Aamir Shafi , Hari Subramoni , Dhabaleswar Panda

Deep neural network training involves both forward propagation (from features through logits to loss) and backward propagation (from loss through gradients to parameter updates). While perturbations along the forward chain, including…

Machine Learning · Computer Science 2026-05-29 Hua Li

A recent line of research has shown that gradient-based algorithms with random initialization can converge to the global minima of the training loss for over-parameterized (i.e., sufficiently wide) deep neural networks. However, the…

Machine Learning · Computer Science 2019-06-12 Difan Zou , Quanquan Gu

Despite the great success of NMT, there still remains a severe challenge: it is hard to interpret the internal dynamics during its training process. In this paper we propose to understand learning dynamics of NMT by using a recent proposed…

Computation and Language · Computer Science 2020-04-07 Conghui Zhu , Guanlin Li , Lemao Liu , Tiejun Zhao , Shuming Shi

Among many mysteries behind the success of deep networks lies the exceptional discriminative power of their learned representations as manifested by the intriguing Neural Collapse (NC) phenomenon, where simple feature structures emerge at…

Machine Learning · Computer Science 2025-10-27 Hancheng Min , Zhihui Zhu , René Vidal

Initially introduced as a machine translation model, the Transformer architecture has now become the foundation for modern deep learning architecture, with applications in a wide range of fields, from computer vision to natural language…

Computation and Language · Computer Science 2024-06-21 Martin Courtois , Malte Ostendorff , Leonhard Hennig , Georg Rehm

We study transformer language models, analyzing attention heads whose attention patterns are spread out, and whose attention scores depend weakly on content. We argue that the softmax denominators of these heads are stable when the…

Computation and Language · Computer Science 2025-10-07 Alex Gibson
‹ Prev 1 8 9 10 Next ›