English
Related papers

Related papers: No Free Swap: Protocol-Dependent Layer Redundancy …

200 papers

Self-supervised pre-training of large-scale transformer models on text corpora followed by finetuning has achieved state-of-the-art on a number of natural language processing tasks. Recently, Lu et al. (2021, arXiv:2103.05247) claimed that…

Machine Learning · Computer Science 2021-07-28 Danielle Rothermel , Margaret Li , Tim Rocktäschel , Jakob Foerster

Deep learning models named transformers achieved state-of-the-art results in a vast majority of NLP tasks at the cost of increased computational complexity and high memory consumption. Using the transformer model in real-time inference…

Final outputs hide when a checkpoint commits to its next-token prediction. We introduce the convergence gap, a model-diffing diagnostic that decodes each layer's next-token distribution and measures its distance to the model's own final…

Machine Learning · Computer Science 2026-05-11 Yifan Zhou

With the rise of large-scale foundation models, efficiently adapting them to downstream tasks remains a central challenge. Linear probing, which freezes the backbone and trains a lightweight head, is computationally efficient but often…

Computer Vision and Pattern Recognition · Computer Science 2026-01-15 Laure Ciernik , Marco Morik , Lukas Thede , Luca Eyring , Shinichi Nakajima , Zeynep Akata , Lukas Muttenthaler

Recently, a test for a sign-changing gap function in a candidate multiband unconventional superconductor involving quasiparticle interference data was proposed. The test was based on the antisymmetric, Fourier transformed conductance maps…

Superconductivity · Physics 2017-05-11 Johannes H. J. Martiny , Andreas Kreisel , P. J. Hirschfeld , Brian M. Andersen

Scaling language models to handle longer contexts introduces substantial memory challenges due to the growing cost of key-value (KV) caches. Motivated by the efficiency gains of hybrid models and the broad availability of pretrained large…

Computation and Language · Computer Science 2026-05-19 Xuan Zhang , Fengzhuo Zhang , Cunxiao Du , Chao Du , Tianyu Pang , Wei Gao , Min Lin

We re-evaluate the standard practice of sharing weights between input and output embeddings in state-of-the-art pre-trained language models. We show that decoupled embeddings provide increased modeling flexibility, allowing us to…

Computation and Language · Computer Science 2020-10-27 Hyung Won Chung , Thibault Févry , Henry Tsai , Melvin Johnson , Sebastian Ruder

A widely cited result by Dong et al. (2021) showed that Transformers built from self-attention alone, without skip connections or feed-forward layers, suffer from rapid rank collapse: all token representations converge to a single…

Machine Learning · Computer Science 2026-04-28 Giansalvo Cirrincione

Gradient-flow analyses show that simplified linear transformers can learn the in-context linear-regression algorithm, but they do not explain the finite-step behavior of gradient descent at large learning rates. Motivated by empirical work…

Machine Learning · Statistics 2026-05-21 Krishnakumar Balasubramanian

Understanding what defines a good representation in large language models (LLMs) is fundamental to both theoretical understanding and practical applications. In this paper, we investigate the quality of intermediate representations in…

Machine Learning · Computer Science 2024-12-13 Oscar Skean , Md Rifat Arefin , Yann LeCun , Ravid Shwartz-Ziv

Leave-One-Out (LOO) provides an intuitive measure of feature importance but is computationally prohibitive. While Layer-Wise Relevance Propagation (LRP) offers a potentially efficient alternative, its axiomatic soundness in modern…

Machine Learning · Computer Science 2025-10-22 Weiqiu You , Siqi Zeng , Yao-Hung Hubert Tsai , Makoto Yamada , Han Zhao

Transformers have revolutionized the field of machine learning. In particular, they can be used to solve complex algorithmic problems, including graph-based tasks. In such algorithmic tasks a key question is what is the minimal size of a…

Machine Learning · Computer Science 2026-01-29 Gilad Yehudai , Clayton Sanford , Maya Bechler-Speicher , Orr Fischer , Ran Gilad-Bachrach , Amir Globerson

Layer-wise normalization (LN) is an essential component of virtually all transformer-based large language models. While its effects on training stability are well documented, its role at inference time is poorly understood. Additionally, LN…

Machine Learning · Computer Science 2025-07-04 Luca Baroni , Galvin Khara , Joachim Schaeffer , Marat Subkhankulov , Stefan Heimersheim

LLM architecture research generally aims to maximize model quality subject to fixed compute/latency budgets. However, many applications of interest such as edge and on-device deployment are further constrained by the model's memory…

Machine Learning · Computer Science 2026-04-28 Abbas Zeitoun , Lucas Torroba-Hennigen , Yoon Kim

In the last decade, Convolutional Neural Network with a multi-layer architecture has advanced rapidly. However, training its complex network is very space-consuming, since a lot of intermediate data are preserved across layers, especially…

Distributed, Parallel, and Cluster Computing · Computer Science 2024-01-23 Zhigang Wang , Hangyu Yang , Ning Wang , Chuanfei Xu , Jie Nie , Zhiqiang Wei , Yu Gu , Ge Yu

We show that MLP layers in transformer language models perform binary routing of continuous signals: the decision of whether a token needs nonlinear processing is well-captured by binary neuron activations, even though the signals being…

Machine Learning · Computer Science 2026-03-12 Peter Balogh

Transformer-based language models create hidden representations of their inputs at every layer, but only use final-layer representations for prediction. This obscures the internal decision-making process of the model and the utility of its…

Computation and Language · Computer Science 2024-06-21 Alexander Yom Din , Taelin Karidi , Leshem Choshen , Mor Geva

In the pursuit of reducing the number of trainable parameters in deep transformer networks, we employ Reinforcement Learning to dynamically select layers during training and tie them together. Every few iterations, the RL agent is asked…

Machine Learning · Computer Science 2024-01-24 Tamir David Hay , Lior Wolf

We investigate multi-scale transformer language models that learn representations of text at multiple scales, and present three different architectures that have an inductive bias to handle the hierarchical nature of language. Experiments…

Computation and Language · Computer Science 2020-05-05 Sandeep Subramanian , Ronan Collobert , Marc'Aurelio Ranzato , Y-Lan Boureau

There remain many open questions pertaining to the scaling behaviour of Transformer architectures. These scaling decisions and findings can be critical, as training runs often come with an associated computational cost which have both…

Computation and Language · Computer Science 2022-02-01 Yi Tay , Mostafa Dehghani , Jinfeng Rao , William Fedus , Samira Abnar , Hyung Won Chung , Sharan Narang , Dani Yogatama , Ashish Vaswani , Donald Metzler