English
Related papers

Related papers: Local to Global: Learning Dynamics and Effect of I…

200 papers

Deep learning involves navigating a high-dimensional loss landscape over the neural network parameter space. Over the course of training, complex computational structures form and re-form inside the neural network, leading to shifts in…

Machine Learning · Computer Science 2025-08-04 Jesse Hoogland , George Wang , Matthew Farrugia-Roberts , Liam Carroll , Susan Wei , Daniel Murfet

Motivated by empirical observations of prolonged plateaus and stage-wise progression during training, we investigate the loss landscape of transformer models trained on in-context next-token prediction tasks. In particular, we focus on…

Machine Learning · Computer Science 2025-08-20 Aditya Varre , Gizem Yüce , Nicolas Flammarion

Transformers have achieved remarkable success in several domains, ranging from natural language processing to computer vision. Nevertheless, it has been recently shown that stacking self-attention layers - the distinctive architectural…

Machine Learning · Computer Science 2022-06-08 Lorenzo Noci , Sotiris Anagnostidis , Luca Biggio , Antonio Orvieto , Sidak Pal Singh , Aurelien Lucchi

The remarkable capability of over-parameterised neural networks to generalise effectively has been explained by invoking a ``simplicity bias'': neural networks prevent overfitting by initially learning simple classifiers before progressing…

Computation and Language · Computer Science 2025-10-02 Riccardo Rende , Federica Gerace , Alessandro Laio , Sebastian Goldt

At present, the mechanisms of in-context learning in Transformers are not well understood and remain mostly an intuition. In this paper, we suggest that training Transformers on auto-regressive objectives is closely related to…

Large Language Models demonstrate remarkable mathematical capabilities but at the same time struggle with abstract reasoning and planning. In this study, we explore whether Transformers can learn to abstract and generalize the rules…

Neural and Evolutionary Computing · Computer Science 2024-12-03 Mikhail Burtsev

Large transformer models pretrained on offline reinforcement learning datasets have demonstrated remarkable in-context reinforcement learning (ICRL) capabilities, where they can make good decisions when prompted with interaction…

Machine Learning · Computer Science 2024-05-28 Licong Lin , Yu Bai , Song Mei

Catastrophic forgetting affects the training of neural networks, limiting their ability to learn multiple tasks sequentially. From the perspective of the well established plasticity-stability dilemma, neural networks tend to be overly…

Machine Learning · Computer Science 2020-06-15 Seyed Iman Mirzadeh , Mehrdad Farajtabar , Razvan Pascanu , Hassan Ghasemzadeh

We study transformers' in-context learning of variable-length Markov chains (VOMCs), focusing on the finite-sample accuracy as the number of in-context examples increases. Compared to fixed-order Markov chains (FOMCs), learning VOMCs is…

Machine Learning · Computer Science 2026-04-01 Ruida Zhou , Chao Tian , Suhas Diggavi

The Transformer is widely used in natural language processing tasks. To train a Transformer however, one usually needs a carefully designed learning rate warm-up stage, which is shown to be crucial to the final performance but will slow…

Machine Learning · Computer Science 2020-06-30 Ruibin Xiong , Yunchang Yang , Di He , Kai Zheng , Shuxin Zheng , Chen Xing , Huishuai Zhang , Yanyan Lan , Liwei Wang , Tie-Yan Liu

We study the problem of length generalization (LG) in transformers: the ability of a model trained on shorter sequences to maintain performance when evaluated on much longer, previously unseen inputs. Prior work by Huang et al. (2025)…

Machine Learning · Computer Science 2025-11-03 Zachary Izzo , Eshaan Nichani , Jason D. Lee

Transformer-based models have achieved remarkable success across a wide range of domains, yet our understanding of their training dynamics remains limited. In this work, we identify a recurrent focus-dilution cycle in attention learning and…

Machine Learning · Computer Science 2026-05-05 Zheng-An Chen , Pengxiao Lin , Zhi-Qin John Xu , Tao Luo

Efficient training and inference algorithms, such as low-rank adaption and model pruning, have shown impressive performance for learning Transformer-based large foundation models. However, due to the technical challenges of the non-convex…

Machine Learning · Computer Science 2024-06-26 Hongkang Li , Meng Wang , Shuai Zhang , Sijia Liu , Pin-Yu Chen

The impact of randomness on model training is poorly understood. How do differences in data order and initialization actually manifest in the model, such that some training runs outperform others or converge faster? Furthermore, how can we…

Machine Learning · Computer Science 2024-01-23 Michael Y. Hu , Angelica Chen , Naomi Saphra , Kyunghyun Cho

In-context learning (ICL) is a hallmark capability of transformers, through which trained models learn to adapt to new tasks by leveraging information from the input context. Prior work has shown that ICL emerges in transformers due to the…

Parametric models, and particularly neural networks, require weight initialization as a starting point for gradient-based optimization. Recent work shows that a specific initial parameter set can be learned from a population of supervised…

Machine Learning · Computer Science 2020-06-12 Lukas Brinkmeyer , Rafael Rego Drumond , Randolf Scholz , Josif Grabocka , Lars Schmidt-Thieme

Transformers have shown a remarkable ability for in-context learning (ICL), making predictions based on contextual examples. However, while theoretical analyses have explored this prediction capability, the nature of the inferred context…

Machine Learning · Computer Science 2025-05-20 Fei Lu , Yue Yu

Despite the empirical success of prompt tuning in adapting pretrained language models to new tasks, theoretical analyses of its capabilities remain limited. Existing theoretical work primarily addresses universal approximation properties,…

Machine Learning · Computer Science 2025-09-03 Maxime Meyer , Mario Michelessa , Caroline Chaux , Vincent Y. F. Tan

Continual learning is the problem of integrating new information in a model while retaining the knowledge acquired in the past. Despite the tangible improvements achieved in recent years, the problem of continual learning is still an open…

Machine Learning · Computer Science 2024-07-24 Giulia Lanzillotta , Sidak Pal Singh , Benjamin F. Grewe , Thomas Hofmann

Transformers, as the fundamental deep learning architecture, have demonstrated great capability in reasoning. This paper studies the generalizable first-order logical reasoning ability of transformers with their parameterized knowledge and…

Computation and Language · Computer Science 2025-07-11 Tianshi Zheng , Jiazheng Wang , Zihao Wang , Jiaxin Bai , Hang Yin , Zheye Deng , Yangqiu Song , Jianxin Li
‹ Prev 1 3 4 5 6 7 10 Next ›