Datasets, Documents, and Repetitions: The Practicalities of Unequal Data Quality

• URL: http://arxiv.org/abs/2503.07879v1
• Date: Mon, 10 Mar 2025 21:51:17 GMT
• Title: Datasets, Documents, and Repetitions: The Practicalities of Unequal Data Quality
• Authors: Alex Fang, Hadi Pouransari, Matt Jordan, Alexander Toshev, Vaishaal Shankar, Ludwig Schmidt, Tom Gunter,
• Abstract summary: We study model performance at various compute budgets and across multiple pre-training datasets created through data filtering and deduplication.<n>We find that, given appropriate modifications to the training recipe, repeating existing aggressively filtered datasets for up to ten epochs can outperform training on the ten times larger superset for a single epoch across multiple compute budget orders of magnitude.
• Score: 67.67387254989018
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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 deduplicated datasets will become a practical constraint. In efforts to better understand how to proceed, we study model performance at various compute budgets and across multiple pre-training datasets created through data filtering and deduplication. We find that, given appropriate modifications to the training recipe, repeating existing aggressively filtered datasets for up to ten epochs can outperform training on the ten times larger superset for a single epoch across multiple compute budget orders of magnitude. While this finding relies on repeating the dataset for many epochs, we also investigate repeats within these datasets at the document level. We find that not all documents within a dataset are equal, and we can create better datasets relative to a token budget by explicitly manipulating the counts of individual documents. We conclude by arguing that even as large language models scale, data filtering remains an important direction of research.

Related papers

• Swift Cross-Dataset Pruning: Enhancing Fine-Tuning Efficiency in Natural Language Understanding [2.379669478864599]
  Current cross-dataset pruning techniques for fine-tuning often rely on computationally expensive sample ranking processes. We propose Swift Cross-Dataset Pruning (SCDP), which uses TF-IDF embeddings with geometric median to rapidly evaluate sample importance. Experimental results on six diverse datasets demonstrate the effectiveness of our method, spanning various tasks and scales.
  arXiv  Detail & Related papers  (2025-01-05T03:52:04Z)
• Scaling Retrieval-Based Language Models with a Trillion-Token Datastore [85.4310806466002]
  We find that increasing the size of the datastore used by a retrieval-based LM monotonically improves language modeling and several downstream tasks without obvious saturation. By plotting compute-optimal scaling curves with varied datastore, model, and pretraining data sizes, we show that using larger datastores can significantly improve model performance for the same training compute budget.
  arXiv  Detail & Related papers  (2024-07-09T08:27:27Z)
• Scaling Laws for Data Filtering -- Data Curation cannot be Compute Agnostic [99.3682210827572]
  Vision-language models (VLMs) are trained for thousands of GPU hours on carefully curated web datasets. Data curation strategies are typically developed agnostic of the available compute for training. We introduce neural scaling laws that account for the non-homogeneous nature of web data.
  arXiv  Detail & Related papers  (2024-04-10T17:27:54Z)
• Data Filtering Networks [67.827994353269]
  We study the problem of learning a data filtering network (DFN) for this second step of filtering a large uncurated dataset. Our key finding is that the quality of a network for filtering is distinct from its performance on downstream tasks. Based on our insights, we construct new data filtering networks that induce state-of-the-art image-text datasets.
  arXiv  Detail & Related papers  (2023-09-29T17:37:29Z)
• Active Data Acquisition in Autonomous Driving Simulation [0.0]
  This paper proposes the concept of an active data-collecting strategy. For high-quality data, increasing the collection density can improve the overall quality of the dataset.
  arXiv  Detail & Related papers  (2023-06-24T10:07:35Z)
• Scaling Data-Constrained Language Models [137.17302576977346]
  We investigate scaling language models in data-constrained regimes. We find that with constrained data for a fixed compute budget, training with up to 4 epochs of repeated data yields negligible changes to loss compared to having unique data. We propose and empirically validate a scaling law for compute optimality that accounts for the decreasing value of repeated tokens and excess parameters.
  arXiv  Detail & Related papers  (2023-05-25T17:18:55Z)
• Optimizing Data Collection for Machine Learning [87.37252958806856]
  Modern deep learning systems require huge data sets to achieve impressive performance. Over-collecting data incurs unnecessary present costs, while under-collecting may incur future costs and delay. We propose a new paradigm for modeling the data collection as a formal optimal data collection problem.
  arXiv  Detail & Related papers  (2022-10-03T21:19:05Z)
• Data Budgeting for Machine Learning [17.524791147624086]
  We study the data budgeting problem and formulate it as two sub-problems. We propose a learning method to solve data budgeting problems. Our empirical evaluation shows that it is possible to perform data budgeting given a small pilot study dataset with as few as $50$ data points.
  arXiv  Detail & Related papers  (2022-10-03T14:53:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.