arXiv:2606. 16112v1 Announce Type: cross Abstract: Residual architectures are ubiquitous in deep learning, but they suffer from a subtle structural limitation: the norm of the residual stream can grow rapidly with depth.
arXiv:2403.04545v4 Announce Type: replace Abstract: Scaling factors in residual branches have emerged as a prevalent method for boosting neural network performance, especially in normalization-free a...
arXiv:2606. 30813v1 Announce Type: cross Abstract: Deep neural networks with repeated architectural blocks, such as transformers, often exhibit structured relationships across layers that emerge during training.
arXiv:2602. 07494v2 Announce Type: replace Abstract: Deeper modern architectures are costly to train, making hyperparameter transfer preferable to expensive repeated tuning.
arXiv:2607. 07884v1 Announce Type: new Abstract: In this short note we consider the gradient descent dynamics of deep scalar linear networks, $f(x) = \prod_{l=1}^L w_l x$, which enjoy exact time-course solutions for any integer depth.
arXiv:2608. 14664v1 Announce Type: new Abstract: How can we determine whether a trained neural network is already deep enough?
arXiv:2602. 05970v2 Announce Type: replace-cross Abstract: Neural scaling laws relate loss to model size in large language models (LLMs), yet depth and width may contribute to performance differently, requiring more detailed studies.
The paper proposes a new strategy for merging layers in deep neural networks, enabling depth compression without requiring an analytical solution for convolutions with padding and without increasing kernel size. This approach addresses limitations of previous methods that struggled with padded convolutions and larger kernels, and it is validated across various architectures and datasets with measured inference speed-ups on embedded platforms.
arXiv:2607. 05017v1 Announce Type: cross Abstract: The performance of deep learning models crucially depends on the settings of hyperparameters like learning rate, initialization scale, and weight decay.
arXiv:2606. 19941v1 Announce Type: new Abstract: Compositionality is believed to be the foundation for generalization, enabling models to reuse meaningful primitives in novel combinations.
arXiv:2606. 26538v1 Announce Type: cross Abstract: Deep Transformers are composed of uniformly stacked residual blocks, yet their deepest layers often add little value.
arXiv:2607. 14018v1 Announce Type: cross Abstract: We investigate how each component of the Transformer feedforward block architecture design determines how much rank survives across depth at initialization.