English
Related papers

Related papers: Perplexity-Aware Data Scaling Law: Perplexity Land…

200 papers

Continual Pre-Training (CPT) has become a popular and effective method to apply strong foundation models to specific downstream tasks. In this work, we explore the learning dynamics throughout the CPT process for large language models. We…

Computation and Language · Computer Science 2025-06-23 Xingjin Wang , Howe Tissue , Lu Wang , Linjing Li , Daniel Dajun Zeng

Continual Pre-Training (CPT) on Large Language Models (LLMs) has been widely used to expand the model's fundamental understanding of specific downstream domains (e.g., math and code). For the CPT on domain-specific LLMs, one important…

Computation and Language · Computer Science 2024-06-04 Haoran Que , Jiaheng Liu , Ge Zhang , Chenchen Zhang , Xingwei Qu , Yinghao Ma , Feiyu Duan , Zhiqi Bai , Jiakai Wang , Yuanxing Zhang , Xu Tan , Jie Fu , Wenbo Su , Jiamang Wang , Lin Qu , Bo Zheng

Large Language Models (LLMs) excel in diverse tasks but often underperform in specialized fields due to limited domain-specific or proprietary corpus. Continual pre-training (CPT) enhances LLM capabilities by imbuing new domain-specific or…

Computation and Language · Computer Science 2024-10-08 Jiawei Gu , Zacc Yang , Chuanghao Ding , Rui Zhao , Fei Tan

Continual pre-training (CPT) for domain adaptation must balance target-domain gains with stability on the base domain. Existing CPT scaling laws typically assume a fixed pre-training budget, which limits their ability to forecast adaptation…

We introduce a framework for optimizing domain-specific dataset construction in foundation model training. Specifically, we seek a cost-efficient way to estimate the quality of data sources (e.g. synthetically generated or filtered web…

Large foundation models are typically trained on data from multiple domains, with the data mixture--the proportion of each domain used--playing a critical role in model performance. The standard approach to selecting this mixture relies on…

Machine Learning · Computer Science 2025-10-03 Mustafa Shukor , Louis Bethune , Dan Busbridge , David Grangier , Enrico Fini , Alaaeldin El-Nouby , Pierre Ablin

In recent years, Large Language Models (LLMs) have made significant strides towards Artificial General Intelligence. However, training these models from scratch requires substantial computational resources and vast amounts of text data. In…

Computation and Language · Computer Science 2024-10-03 Wenzhen Zheng , Wenbo Pan , Xu Xu , Libo Qin , Li Yue , Ming Zhou

Conventional continual pretraining (CPT) for large language model (LLM) domain adaptation often suffers from catastrophic forgetting and limited domain capacity. Existing strategies adopt layer expansion, introducing additional trainable…

Machine Learning · Computer Science 2025-10-14 Jinyang Zhang , Yue Fang , Hongxin Ding , Weibin Liao , Muyang Ye , Xu Chu , Junfeng Zhao , Yasha Wang

Quality pretraining data is often seen as the key to high-performance language models. However, progress in understanding pretraining data has been slow due to the costly pretraining runs required for data selection experiments. We present…

Computation and Language · Computer Science 2025-03-11 Tristan Thrush , Christopher Potts , Tatsunori Hashimoto

Continual Pre-Training (CPT) is widely used for acquiring and updating factual knowledge in LLMs. This practice treats loss as a proxy for knowledge learning, while offering no grounding into how it changes during training. We study CPT as…

Computation and Language · Computer Science 2026-01-08 Seyed Mahed Mousavi , Simone Alghisi , Giuseppe Riccardi

Continued pretraining (CPT) is a popular approach to adapt existing large language models (LLMs) to new languages. When doing so, it is common practice to include a portion of English data in the mixture, but its role has not been carefully…

Computation and Language · Computer Science 2025-09-22 Ahmed Elhady , Eneko Agirre , Mikel Artetxe

Precise estimation of downstream performance in large language models (LLMs) prior to training is essential for guiding their development process. Scaling laws analysis utilizes the statistics of a series of significantly smaller sampling…

Computation and Language · Computer Science 2025-04-09 Yangyi Chen , Binxuan Huang , Yifan Gao , Zhengyang Wang , Jingfeng Yang , Heng Ji

Neural scaling laws define a predictable relationship between a model's parameter count and its performance after training in the form of a power law. However, most research to date has not explicitly investigated whether scaling laws can…

Computation and Language · Computer Science 2022-10-19 Maor Ivgi , Yair Carmon , Jonathan Berant

While metrics available during pre-training, such as perplexity, correlate well with model performance at scaling-laws studies, their predictive capacities at a fixed model size remain unclear, hindering effective model selection and…

Computation and Language · Computer Science 2025-10-17 Hansi Zeng , Kai Hui , Honglei Zhuang , Zhen Qin , Zhenrui Yue , Hamed Zamani , Dana Alon

Traditional scaling laws in natural language processing suggest that increasing model size and training data enhances performance. However, recent studies reveal deviations, particularly in large language models, where performance…

Machine Learning · Computer Science 2025-07-16 Zhengyu Chen , Siqi Wang , Teng Xiao , Yudong Wang , Shiqi Chen , Xunliang Cai , Junxian He , Jingang Wang

Upweighting high-quality data in LLM pretraining often improves performance, but in datalimited regimes, especially under overtraining, stronger upweighting increases repetition and can degrade performance. However, standard scaling laws do…

Computation and Language · Computer Science 2026-05-05 Fengze Liu , Weidong Zhou , Binbin Liu , Ping Guo , Zijun Wang , Bingni Zhang , Yifan Zhang , Yifeng Yu , Xiaohuan Zhou , Taifeng Wang

Scaling laws for language model training traditionally characterize how performance scales with model size and dataset volume. Prior work has explored architecture variants and data treatments such as dataset filtering and noise injection…

Machine Learning · Computer Science 2026-02-24 Anirudh Subramanyam , Yuxin Chen , Robert L. Grossman

Large volumes of text data have contributed significantly to the development of large language models (LLMs) in recent years. This data is typically acquired by scraping the internet, leading to pretraining datasets comprised of noisy web…

Computation and Language · Computer Science 2023-09-12 Max Marion , Ahmet Üstün , Luiza Pozzobon , Alex Wang , Marzieh Fadaee , Sara Hooker

Low precision training and inference affect both the quality and cost of language models, but current scaling laws do not account for this. In this work, we devise "precision-aware" scaling laws for both training and inference. We propose…

Continual learning (CL) in large language models (LLMs) is an evolving domain that focuses on developing efficient and sustainable training strategies to adapt models to emerging knowledge and achieve robustness in dynamic environments. Our…

Computation and Language · Computer Science 2025-02-13 Çağatay Yıldız , Nishaanth Kanna Ravichandran , Nitin Sharma , Matthias Bethge , Beyza Ermis
‹ Prev 1 2 3 10 Next ›