paper-with-me

Papers

Learning Rate Transfer in Normalized Transformers

2026-04-29 · Boris Shigida, Boris Hanin, Andrey Gromov arxiv

The Normalized Transformer, or nGPT (arXiv:2410.01131) achieves impressive training speedups and does not require weight decay or learning rate warmup. However, despite having hyperparameters that explicitly scale with model size, we observe that nGPT does not exhibit learning rate transfer across model dimension and token horizon. To rectify this, we combine numerical experiments with a principled use of alignment exponents (arXiv:2407.05872) to revisit and modify the $μ$P approach to hyperparameter transfer (arXiv:2011.14522). The result is a novel nGPT parameterization we call $ν$GPT. Through extensive empirical validation, we find $ν$GPT exhibits learning rate transfer across width, depth, and token horizon.

📄 PDF Abstract BibTeX arXiv:2604.27077

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

What Dense Graph Do You Need for Self-Attention?

2022-05-27 · Yuxin Wang, Chu-Tak Lee, Qipeng Guo, Zhangyue Yin 외

Transformers have made progress in miscellaneous tasks, but suffer from quadratic computational and memory complexities. Recent works propose sparse Transformers with attention on sparse graphs to reduce complexity and r…

Miscellaneous

Normalize Filters! Classical Wisdom for Deep Vision

2025-06-04 · Gustavo Perez, Stella X. Yu

Classical image filters, such as those for averaging or differencing, are carefully normalized to ensure consistency, interpretability, and to avoid artifacts like intensity shifts, halos, or ringing. In contrast, convol…

Transformers Efficiently Perform In-Context Logistic Regression via Normalized Gradient Descent

2026-05-07 · Chenyang Zhang, Yuan Cao arxiv

Transformers have demonstrated remarkable in-context learning (ICL) capabilities. The strong ICL performance of transformers is commonly believed to arise from their ability to implicitly execute certain algorithms on th…

OrbitQuant: Data-Agnostic Quantization for Image and Video Diffusion Transformers

2026-07-02 · Donghyun Lee, Jitesh Chavan, Duy Nguyen, Sam Huang 외 arxiv

Diffusion transformers (DiTs) achieve state-of-the-art image and video generation, but their multi-step sampling and growing parameter count make inference expensive. Post-training quantization (PTQ) is the natural remed…

Video Generation

DNT: a Deeply Normalized Transformer that can be trained by Momentum SGD

2025-07-23 · Xianbiao Qi, Marco Chen, Wenjie Xiao, Jiaquan Ye 외 arxiv

Transformers have become the de facto backbone of modern deep learning, yet their training typically demands an advanced optimizer with adaptive learning rate like AdamW, rather than a momentum SGDW (mSGDW). Previous wor…