English
Related papers

Related papers: Shaping capabilities with token-level data filteri…

200 papers

Token filtering to reduce irrelevant tokens prior to self-attention is a straightforward way to enable efficient vision Transformer. This is the first work to view token filtering from a feature selection perspective, where we weigh the…

Computer Vision and Pattern Recognition · Computer Science 2023-05-25 Hong Wang , Su Yang , Xiaoke Huang , Weishan Zhang

Recent studies show that in supervised fine-tuning (SFT) of large language models (LLMs), data quality matters more than quantity. While most data cleaning methods concentrate on filtering entire samples, the quality of individual tokens…

Computation and Language · Computer Science 2026-03-12 Jinlong Pang , Na Di , Zhaowei Zhu , Jiaheng Wei , Hao Cheng , Chen Qian , Yang Liu

Pretraining is the preliminary and fundamental step in developing capable language models (LM). Despite this, pretraining data design is critically under-documented and often guided by empirically unsupported intuitions. To address this, we…

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

Data filtering has become a powerful tool for improving model performance while reducing computational cost. However, as large language model compute budgets continue to grow, the limited data volume provided by heavily filtered and…

Computation and Language · Computer Science 2025-11-07 Alex Fang , Hadi Pouransari , Matt Jordan , Alexander Toshev , Vaishaal Shankar , Ludwig Schmidt , Tom Gunter

Transformers achieve unrivalled performance in modelling language, but remain inefficient in terms of memory and time complexity. A possible remedy is to reduce the sequence length in the intermediate layers by pooling fixed-length segments…

Computation and Language · Computer Science 2023-10-25 Piotr Nawrot , Jan Chorowski , Adrian Łańcucki , Edoardo M. Ponti

Pretrained transformer models have achieved state-of-the-art results in many tasks and benchmarks recently. Many state-of-the-art Language Models (LMs), however, do not scale well above the threshold of 512 input tokens. In specialized…

Computation and Language · Computer Science 2022-12-01 Joel Niklaus , Daniele Giofré

Recently, pre-trained language models mostly follow the pre-train-then-fine-tuning paradigm and have achieved great performance on various downstream tasks. However, since the pre-training stage is typically task-agnostic and the…

Computation and Language · Computer Science 2020-10-08 Yuxian Gu , Zhengyan Zhang , Xiaozhi Wang , Zhiyuan Liu , Maosong Sun

As large language models (LLMs) are pretrained on massive web corpora, careful selection of data becomes essential to ensure effective and efficient learning. While perplexity (PPL)-based filtering has shown strong performance, it suffers…

Computation and Language · Computer Science 2026-03-04 Yeongbin Seo , Gayoung Kim , Jaehyung Kim , Jinyoung Yeo

Transformer-based models generally allocate the same amount of computation for each token in a given sequence. We develop a simple but effective "token dropping" method to accelerate the pretraining of transformer models, such as BERT,…

Computation and Language · Computer Science 2022-03-25 Le Hou , Richard Yuanzhe Pang , Tianyi Zhou , Yuexin Wu , Xinying Song , Xiaodan Song , Denny Zhou

Instruction tuning is critical to improve LLMs but usually suffers from low-quality and redundant data. Data filtering for instruction tuning has proved important in improving both the efficiency and performance of the tuning process. But…

Computation and Language · Computer Science 2024-06-11 Ming Li , Yong Zhang , Shwai He , Zhitao Li , Hongyu Zhao , Jianzong Wang , Ning Cheng , Tianyi Zhou

Continued pre-training of small language models offers a promising path for domain adaptation with limited computational resources. I've investigated this approach within educational domains, evaluating it as a resource-efficient…

Computation and Language · Computer Science 2025-04-15 Salman Faroz

Token dropping is a recently-proposed strategy to speed up the pretraining of masked language models, such as BERT, by skipping the computation of a subset of the input tokens at several middle layers. It can effectively reduce the training…

Computation and Language · Computer Science 2023-05-25 Qihuang Zhong , Liang Ding , Juhua Liu , Xuebo Liu , Min Zhang , Bo Du , Dacheng Tao

Large-scale models are pretrained on massive web-crawled datasets containing documents of mixed quality, making data filtering essential. A popular method is Classifier-based Quality Filtering (CQF), which trains a binary classifier to…

Machine Learning · Computer Science 2025-10-03 Thiziri Nait Saada , Louis Bethune , Michal Klein , David Grangier , Marco Cuturi , Pierre Ablin

Large language models pretrained on general-domain corpora often exhibit tokenization inefficiencies when applied to specialized domains. Although continual pretraining for domain adaptation partially alleviate performance degradation, it…

Computation and Language · Computer Science 2026-05-19 Gunjan Balde , Soumyadeep Roy , Mainack Mondal , Niloy Ganguly

Current tokenization methods process sequential data without accounting for signal quality, limiting their effectiveness on noisy real-world corpora. We present QA-Token (Quality-Aware Tokenization), which incorporates data reliability…

Artificial Intelligence · Computer Science 2026-02-09 Arvid E. Gollwitzer , Paridhi Latawa , David de Gruijl , Deepak A. Subramanian , Adrián Noriega de la Colina

Improvements in language model capabilities are often attributed to increasing model size or training data, but in some cases smaller models trained on curated data or with different architectural decisions can outperform larger ones…

Curriculum learning-organizing training data from easy to hard-has improved efficiency across machine learning domains, yet remains underexplored for language model pretraining. We present the first systematic investigation of curriculum…

Computation and Language · Computer Science 2026-01-29 Yang Zhang , Amr Mohamed , Hadi Abdine , Guokan Shang , Michalis Vazirgiannis

The pre-training of masked language models (MLMs) consumes massive computation to achieve good results on downstream NLP tasks, resulting in a large carbon footprint. In the vanilla MLM, the virtual tokens, [MASK]s, act as placeholders and…

Computation and Language · Computer Science 2022-11-16 Baohao Liao , David Thulke , Sanjika Hewavitharana , Hermann Ney , Christof Monz

State-of-the-art pre-trained language models have been shown to memorise facts and perform well with limited amounts of training data. To gain a better understanding of how these models learn, we study their generalisation and memorisation…

Computation and Language · Computer Science 2022-03-16 Michael Tänzer , Sebastian Ruder , Marek Rei
‹ Prev 1 2 3 10 Next ›