English
Related papers

Related papers: Reusing Overtrained Language Models Saturates Scal…

200 papers

Scaling laws in language modeling traditionally quantify training loss as a function of dataset size and model parameters, providing compute-optimal estimates but often neglecting the impact of data quality on model generalization. In this…

Computation and Language · Computer Science 2024-10-07 Ernie Chang , Matteo Paltenghi , Yang Li , Pin-Jie Lin , Changsheng Zhao , Patrick Huber , Zechun Liu , Rastislav Rabatin , Yangyang Shi , Vikas Chandra

We classify and re-examine some of the current approaches to improve the performance-computes trade-off of language models, including (1) non-causal models (such as masked language models), (2) extension of batch length with efficient…

Computation and Language · Computer Science 2020-09-16 Aran Komatsuzaki

The pre-training phase of language models often begins with randomly initialized parameters. With the current trends in scaling models, training their large number of parameters can be extremely slow and costly. In contrast, small language…

Computation and Language · Computer Science 2024-09-23 Mohammad Samragh , Iman Mirzadeh , Keivan Alizadeh Vahid , Fartash Faghri , Minsik Cho , Moin Nabi , Devang Naik , Mehrdad Farajtabar

While scaling laws for large language models (LLMs) during pre-training have been extensively studied, their behavior under reinforcement learning (RL) post-training remains largely unexplored. This paper presents a systematic empirical…

Large language models (LLMs) are routinely pre-trained on billions of tokens, only to start the process over again once new data becomes available. A much more efficient solution is to continually pre-train these models, saving significant…

Tokenization and transfer learning are two critical components in building state of the art time series foundation models for forecasting. In this work, we systematically study the effect of tokenizer design, specifically scaling and…

Machine Learning · Computer Science 2025-11-18 Alexis Roger , Gwen Legate , Kashif Rasul , Yuriy Nevmyvaka , Irina Rish

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

While scaling laws for Large Language Models (LLMs) traditionally focus on proxy metrics like pretraining loss, predicting downstream task performance has been considered unreliable. This paper challenges that view by proposing a direct…

Machine Learning · Computer Science 2025-12-10 Jakub Krajewski , Amitis Shidani , Dan Busbridge , Sam Wiseman , Jason Ramapuram

Training large models from scratch usually costs a substantial amount of resources. Towards this problem, recent studies such as bert2BERT and LiGO have reused small pretrained models to initialize a large model (termed the ``target…

Machine Learning · Computer Science 2023-10-18 Yu Pan , Ye Yuan , Yichun Yin , Zenglin Xu , Lifeng Shang , Xin Jiang , Qun Liu

Retrieval-augmented generation (RAG) improves language model (LM) performance by providing relevant context at test time for knowledge-intensive situations. However, the relationship between parametric knowledge acquired during pretraining…

Computation and Language · Computer Science 2026-04-02 Karan Singh , Michael Yu , Varun Gangal , Zhuofu Tao , Sachin Kumar , Emmy Liu , Steven Y. Feng

As text and code resources have expanded, large-scale pre-trained models have shown promising capabilities in code generation tasks, typically employing supervised fine-tuning with problem statement-program pairs. However, increasing model…

Computation and Language · Computer Science 2025-04-10 Nathanaël Beau , Benoît Crabbé

Modern language models rely on static vocabularies, fixed before pretraining, in contrast to the adaptive vocabulary acquisition observed in human language learning. To bridge this gap, we introduce vocabulary curriculum learning, an…

Computation and Language · Computer Science 2025-02-26 Fangyuan Yu

Pretraining large language models (LLMs) is resource-intensive, often requiring months of training time even with high-end GPU clusters. There are two approaches of mitigating such computational demands: reusing smaller models to train…

Machine Learning · Computer Science 2025-06-17 Seng Pei Liew , Takuya Kato , Sho Takase

Most languages lack sufficient data for large-scale monolingual pretraining, creating a "data wall." Multilingual pretraining helps but is limited by language imbalance and the "curse of multilinguality." An alternative is to translate…

Computation and Language · Computer Science 2025-09-23 Dan John Velasco , Matthew Theodore Roque

Pretraining NLP models with variants of Masked Language Model (MLM) objectives has recently led to a significant improvements on many tasks. This paper examines the benefits of pretrained models as a function of the number of training…

Computation and Language · Computer Science 2020-06-17 Sinong Wang , Madian Khabsa , Hao Ma

Generative recommendation models can model user behavior as sequences of events and provide a shared backbone for multiple recommendation tasks. In production, however, pre-training gains do not automatically translate into downstream…

Information Retrieval · Computer Science 2026-05-25 Qiuling Xu , Ko-Jen Hsiao , Moumita Bhattacharya

Large language models have led to state-of-the-art accuracies across a range of tasks. However,training large language model needs massive computing resource, as more and more open source pre-training models are available, it is worthy to…

Computation and Language · Computer Science 2021-04-26 Han Zhang

Language models have demonstrated remarkable performance in solving reasoning tasks; however, even the strongest models still occasionally make reasoning mistakes. Recently, there has been active research aimed at improving reasoning…

Computation and Language · Computer Science 2024-08-30 Tian Ye , Zicheng Xu , Yuanzhi Li , Zeyuan Allen-Zhu

Scaling laws predict the loss of a target machine learning model by extrapolating from easier-to-train models with fewer parameters or smaller training sets. This provides an efficient way for practitioners and researchers alike to compare…

Machine Learning · Computer Science 2025-06-04 Leshem Choshen , Yang Zhang , Jacob Andreas

Tokenization is a fundamental component of large language models (LLMs), yet its influence on model scaling and performance is not fully explored. In this paper, we introduce Over-Tokenized Transformers, a novel framework that decouples…

Computation and Language · Computer Science 2025-05-26 Hongzhi Huang , Defa Zhu , Banggu Wu , Yutao Zeng , Ya Wang , Qiyang Min , Xun Zhou