English
Related papers

Related papers: Curriculum Learning for LLM Pretraining: An Analys…

200 papers

In humans and animals, curriculum learning -- presenting data in a curated order - is critical to rapid learning and effective pedagogy. Yet in machine learning, curricula are not widely used and empirically often yield only moderate…

Machine Learning · Computer Science 2022-12-07 Luca Saglietti , Stefano Sarao Mannelli , Andrew Saxe

It is common knowledge that the quantity and quality of the training data play a significant role in the creation of a good machine learning model. In this paper, we take it one step further and demonstrate that the way the training…

Audio and Speech Processing · Electrical Eng. & Systems 2022-08-12 Georgios Karakasidis , Tamás Grósz , Mikko Kurimo

Machine translation systems based on deep neural networks are expensive to train. Curriculum learning aims to address this issue by choosing the order in which samples are presented during training to help train better models faster. We…

Computation and Language · Computer Science 2018-11-05 Xuan Zhang , Gaurav Kumar , Huda Khayrallah , Kenton Murray , Jeremy Gwinnup , Marianna J Martindale , Paul McNamee , Kevin Duh , Marine Carpuat

Language Models like ELMo and BERT have provided robust representations of natural language, which serve as the language understanding component for a diverse range of downstream tasks.Curriculum learning is a method that employs a…

Computation and Language · Computer Science 2021-08-05 Daniel Campos

Curriculum Learning (CL) is a technique of training models via ranking examples in a typically increasing difficulty trend with the aim of accelerating convergence and improving generalisability. Current approaches for Natural Language…

Computation and Language · Computer Science 2022-11-28 Fenia Christopoulou , Gerasimos Lampouras , Ignacio Iacobacci

Curriculum learning has been used to improve the quality of text generation systems by ordering the training samples according to a particular schedule in various tasks. In the context of data-to-text generation (DTG), previous studies used…

Computation and Language · Computer Science 2024-12-19 Kancharla Aditya Hari , Manish Gupta , Vasudeva Varma

Neural ranking models are traditionally trained on a series of random batches, sampled uniformly from the entire training set. Curriculum learning has recently been shown to improve neural models' effectiveness by sampling batches…

Information Retrieval · Computer Science 2019-12-19 Gustavo Penha , Claudia Hauff

The structure of curriculum plays a vital role in our learning process, both as children and adults. Presenting material in ascending order of difficulty that also exploits prior knowledge can have a significant impact on the rate of…

Computation and Language · Computer Science 2017-11-30 Ahmed H. Zaidi , Russell Moore , Ted Briscoe

Web-scale pre-training datasets are the cornerstone of LLMs' success. However, text data curated from the Internet inevitably contains random noise caused by decoding errors or unregulated web content. In contrast to previous works that…

Computation and Language · Computer Science 2025-05-19 Jinghan Ru , Yuxin Xie , Xianwei Zhuang , Yuguo Yin , Zhihui Guo , Zhiming Liu , Qianli Ren , Yuexian Zou

When random label noise is added to a training dataset, the prediction error of a neural network on a label-noise-free test dataset initially improves during early training but eventually deteriorates, following a U-shaped dependence on…

Machine Learning · Computer Science 2023-06-06 Chaoyue Liu , Amirhesam Abedsoltan , Mikhail Belkin

Large language models (LLMs) generally utilize a consistent data distribution throughout the pretraining process. However, as the model's capability improves, it is intuitive that its data preferences dynamically change, indicating the need…

Computation and Language · Computer Science 2025-02-18 Xuemiao Zhang , Liangyu Xu , Feiyu Duan , Yongwei Zhou , Sirui Wang , Rongxiang Weng , Jingang Wang , Xunliang Cai

The order of training samples plays a crucial role in large language models (LLMs), significantly impacting both their external performance and internal learning dynamics. Traditional methods for investigating this effect generally require…

Machine Learning · Computer Science 2025-05-29 Hao Yang , Haoxuan Li , Mengyue Yang , Xu Chen , Mingming Gong

Large-scale pretraining datasets drive the success of large language models (LLMs). However, these web-scale corpora inevitably contain large amounts of noisy data due to unregulated web content or randomness inherent in data. Although LLM…

Machine Learning · Computer Science 2026-02-03 Qizhen Zhang , Ankush Garg , Jakob Foerster , Niladri Chatterji , Kshitiz Malik , Mike Lewis

Chain-of-thought reasoning, where language models expend additional computation by producing thinking tokens prior to final responses, has driven significant advances in model capabilities. However, training these reasoning models is…

Machine Learning · Computer Science 2026-03-20 Nived Rajaraman , Audrey Huang , Miro Dudik , Robert Schapire , Dylan J. Foster , Akshay Krishnamurthy

We use Bayesian optimization to learn curricula for word representation learning, optimizing performance on downstream tasks that depend on the learned representations as features. The curricula are modeled by a linear ranking function…

Computation and Language · Computer Science 2016-06-22 Yulia Tsvetkov , Manaal Faruqui , Wang Ling , Brian MacWhinney , Chris Dyer

We introduce a method for automatically selecting the path, or syllabus, that a neural network follows through a curriculum so as to maximise learning efficiency. A measure of the amount that the network learns from each data sample is…

Neural and Evolutionary Computing · Computer Science 2017-04-12 Alex Graves , Marc G. Bellemare , Jacob Menick , Remi Munos , Koray Kavukcuoglu

Most curriculum learning methods require an approach to sort the data samples by difficulty, which is often cumbersome to perform. In this work, we propose a novel curriculum learning approach termed Learning Rate Curriculum (LeRaC), which…

Machine Learning · Computer Science 2024-07-23 Florinel-Alin Croitoru , Nicolae-Catalin Ristea , Radu Tudor Ionescu , Nicu Sebe

We propose a learning approach for turn-level spoken language understanding, which facilitates a user to speak one or more utterances compositionally in a turn for completing a task (e.g., voice ordering). A typical pipelined approach for…

Computation and Language · Computer Science 2019-06-12 Hao Lang , Wen Wang

We show that language models' activations linearly encode when information was learned during training. Our setup involves creating a model with a known training order by sequentially fine-tuning Llama-3.2-1B on six disjoint but otherwise…

Machine Learning · Computer Science 2025-09-23 Dmitrii Krasheninnikov , Richard E. Turner , David Krueger

Pretraining large language models effectively requires strategic data selection, blending and ordering. However, key details about data mixtures especially their scalability to longer token horizons and larger model sizes remain…

Computation and Language · Computer Science 2024-12-23 Steven Feng , Shrimai Prabhumoye , Kezhi Kong , Dan Su , Mostofa Patwary , Mohammad Shoeybi , Bryan Catanzaro