English
Related papers

Related papers: Taming Transformer Without Using Learning Rate War…

200 papers

We introduce directional routing, a lightweight mechanism that gives each transformer attention head learned suppression directions controlled by a shared router, at 3.9% parameter cost. We train a 433M-parameter model alongside an…

Machine Learning · Computer Science 2026-03-17 Kevin Taylor

Massive transformer-based models face several challenges, including slow and computationally intensive pre-training and over-parametrization. This paper addresses these challenges by proposing a versatile method called GQKVA, which…

Commonly used optimization algorithms often show a trade-off between good generalization and fast training times. For instance, stochastic gradient descent (SGD) tends to have good generalization; however, adaptive gradient methods have…

Machine Learning · Computer Science 2023-06-14 Aditya Cowsik , Tankut Can , Paolo Glorioso

The quadratic computational cost of the self-attention mechanism is a primary challenge in scaling Transformer models. While attention sparsity is widely studied as a technique to improve computational efficiency, it is almost universally…

Computation and Language · Computer Science 2025-08-11 Sagar Gandhi , Vishal Gandhi

When incrementally trained on new classes, deep neural networks are subject to catastrophic forgetting which leads to an extreme deterioration of their performance on the old classes while learning the new ones. Using a small memory…

Machine Learning · Computer Science 2022-11-15 Quentin Jodelet , Xin Liu , Tsuyoshi Murata

One of the main challenges in optimal scaling of large language models (LLMs) is the prohibitive cost of hyperparameter tuning, particularly learning rate $\eta$ and batch size $B$. While techniques like $\mu$P (Yang et al., 2022) provide…

Machine Learning · Computer Science 2025-01-10 Oleg Filatov , Jan Ebert , Jiangtao Wang , Stefan Kesselheim

Emergence is a fascinating property of large language models and neural networks more broadly: as models scale and train for longer, they sometimes develop new abilities in sudden ways. Despite initial studies, we still lack a comprehensive…

Machine Learning · Computer Science 2025-12-11 Nicolas Zucchet , Francesco d'Angelo , Andrew K. Lampinen , Stephanie C. Y. Chan

Adapting pre-trained models to specialized tasks often leads to catastrophic forgetting, where new knowledge overwrites foundational capabilities. Existing methods either compromise performance on the new task or struggle to balance…

Machine Learning · Computer Science 2026-03-10 Dyah Adila , Hanna Mazzawi , Benoit Dherin , Xavier Gonzalvo

Embedding tables are usually huge in click-through rate (CTR) prediction models. To train and deploy the CTR models efficiently and economically, it is necessary to compress their embedding tables at the training stage. To this end, we…

Machine Learning · Computer Science 2024-08-07 Shiwei Li , Huifeng Guo , Lu Hou , Wei Zhang , Xing Tang , Ruiming Tang , Rui Zhang , Ruixuan Li

As a foundational architecture of artificial intelligence models, Transformer has been recently adapted to spiking neural networks with promising performance across various tasks. However, existing spiking Transformer(ST)-based models…

Machine Learning · Computer Science 2026-01-06 Hongze Sun , Wuque Cai , Duo Chen , Quan Tang , Shifeng Mao , Jiayi He , Zhenxing Wang , Yan Cui , Dezhong Yao , Daqing Guo

Trained models are often composed with post-hoc transforms such as temperature scaling (TS), ensembling and stochastic weight averaging (SWA) to improve performance, robustness, uncertainty estimation, etc. However, such transforms are…

Machine Learning · Computer Science 2024-10-07 Rishabh Ranjan , Saurabh Garg , Mrigank Raman , Carlos Guestrin , Zachary Lipton

Training large transformer models is one of the most important computational challenges of modern AI. In this paper, we show how to significantly accelerate training of large transformer models by reducing activation recomputation.…

Machine Learning · Computer Science 2022-05-12 Vijay Korthikanti , Jared Casper , Sangkug Lym , Lawrence McAfee , Michael Andersch , Mohammad Shoeybi , Bryan Catanzaro

Recently, Transformer-based language models have demonstrated remarkable performance across many NLP domains. However, the unsupervised pre-training step of these models suffers from unbearable overall computational expenses. Current…

Machine Learning · Computer Science 2020-10-27 Minjia Zhang , Yuxiong He

Training instability remains a critical challenge in large language model (LLM) pretraining, often manifesting as sudden gradient explosions that waste significant computational resources. We study training failures in a 5M-parameter…

Machine Learning · Computer Science 2026-02-03 Lianhai Ren , Yucheng Ding , Xiao Liu , Qianxiao Li , Peng Cheng , Yeyun Gong

Training deep neural networks using a large batch size has shown promising results and benefits many real-world applications. However, the optimizer converges slowly at early epochs and there is a gap between large-batch deep learning…

Machine Learning · Computer Science 2020-02-06 Zhouyuan Huo , Bin Gu , Heng Huang

This paper proposes a novel, efficient transfer learning method, called Scalable Weight Reparametrization (SWR) that is efficient and effective for multiple downstream tasks. Efficient transfer learning involves utilizing a pre-trained…

Machine Learning · Computer Science 2023-02-28 Byeonggeun Kim , Jun-Tae Lee , Seunghan yang , Simyung Chang

Large Transformer-based language models are pre-trained on corpora of varying sizes, for a different number of steps and with different batch sizes. At the same time, more fundamental components, such as the pre-training objective or…

Computation and Language · Computer Science 2021-05-12 M. Aßenmacher , P. Schulze , C. Heumann

Reinforcement learning with verifiable rewards (RLVR) has emerged as a prominent paradigm for enhancing the reasoning capabilities of large language models (LLMs). However, the entropy of LLMs usually collapses during RLVR training, leading…

Computation and Language · Computer Science 2026-04-21 Renren Jin , Pengzhi Gao , Yuqi Ren , Zhuowen Han , Tongxuan Zhang , Wuwei Huang , Wei Liu , Jian Luan , Deyi Xiong

As hyperparameter tuning becomes increasingly costly at scale, efficient tuning methods are essential. Yet principles for guiding hyperparameter tuning remain limited. In this work, we seek to establish such principles by considering a…

Machine Learning · Computer Science 2025-09-30 Bingrui Li , Jiaxin Wen , Zhanpeng Zhou , Jun Zhu , Jianfei Chen

With the popularity of the recent Transformer-based models represented by BERT, GPT-3 and ChatGPT, there has been state-of-the-art performance in a range of natural language processing tasks. However, the massive computations, huge memory…

Computation and Language · Computer Science 2023-04-04 Gaochen Dong , Wei Chen