Is More Data Worth the Cost? Dataset Scaling Laws in a Tiny Attention-Only Decoder

• URL: http://arxiv.org/abs/2604.09389v1
• Date: Fri, 10 Apr 2026 15:02:59 GMT
• Title: Is More Data Worth the Cost? Dataset Scaling Laws in a Tiny Attention-Only Decoder
• Abstract summary: Training Transformer language models are expensive, as performance typically improves with increasing dataset size and computational budget.<n>In this work, we isolate dataset-size effects using a strongly reduced attention-only decoder architecture.<n>We observe smooth performance improvements accompanied by clear diminishing returns, consistent with scaling-law behavior.
• Score: 1.697082719450444
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Training Transformer language models is expensive, as performance typically improves with increasing dataset size and computational budget. Although scaling laws describe this trend at large scale, their implications in controlled, smaller-scale settings remain less explored. In this work, we isolate dataset-size effects using a strongly reduced attention-only decoder architecture. By training on progressively larger power-of-two subsets, we observe smooth performance improvements accompanied by clear diminishing returns, consistent with scaling-law behavior. Using only about 30% of the training data is sufficient to reach approximately 90% of the full-data validation token-level accuracy. These results provide actionable insights into dataset scaling in a controlled, component-isolated setting and offer practical guidance for balancing dataset size and computational cost in compute- and data-restricted environments, such as small research labs and exploratory model development.

Related papers

• LLMs on the Line: Data Determines Loss-to-Loss Scaling Laws [21.053622641336744]
  Loss-to-loss scaling laws relate losses across pretraining datasets and downstream tasks.<n>Our experiments reveal that the pretraining data and tokenizer determine the scaling trend.
  arXiv  Detail & Related papers  (2025-02-17T18:45:25Z)
• 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 the Value of Individual Data Points in Machine Learning [55.596413470429475]
  We introduce a new perspective by investigating scaling behavior for the value of individual data points. We provide learning theory to support our scaling law, and we observe empirically that it holds across diverse model classes. Our work represents a first step towards understanding and utilizing scaling properties for the value of individual data points.
  arXiv  Detail & Related papers  (2024-05-30T20:10:24Z)
• An In-Depth Analysis of Data Reduction Methods for Sustainable Deep Learning [0.15833270109954137]
  We present up to eight different methods to reduce the size of a training dataset. We also develop a Python package to apply them. We experimentally compare how these data reduction methods affect the representativeness of the reduced dataset.
  arXiv  Detail & Related papers  (2024-03-22T12:06:40Z)
• Scaling Laws for Sparsely-Connected Foundation Models [70.41266138010657]
  We explore the impact of parameter sparsity on the scaling behavior of Transformers trained on massive datasets. We identify the first scaling law describing the relationship between weight sparsity, number of non-zero parameters, and amount of training data.
  arXiv  Detail & Related papers  (2023-09-15T16:29:27Z)
• Scaling Data-Constrained Language Models [133.2083255645999]
  We investigate scaling language models in data-constrained regimes.<n>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.<n>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)
• How Much More Data Do I Need? Estimating Requirements for Downstream Tasks [99.44608160188905]
  Given a small training data set and a learning algorithm, how much more data is necessary to reach a target validation or test performance? Overestimating or underestimating data requirements incurs substantial costs that could be avoided with an adequate budget. Using our guidelines, practitioners can accurately estimate data requirements of machine learning systems to gain savings in both development time and data acquisition costs.
  arXiv  Detail & Related papers  (2022-07-04T21:16:05Z)
• Beyond neural scaling laws: beating power law scaling via data pruning [37.804100045519846]
  We show how in theory we can break beyond power law scaling and potentially even reduce it to exponential scaling. We develop a new simple, cheap and scalable self-supervised pruning metric that demonstrates comparable performance to the best supervised metrics. Overall, our work suggests that the discovery of good data-pruning metrics may provide a viable path forward to substantially improved neural scaling laws.
  arXiv  Detail & Related papers  (2022-06-29T09:20:47Z)
• Data Scaling Laws in NMT: The Effect of Noise and Architecture [59.767899982937756]
  We study the effect of varying the architecture and training data quality on the data scaling properties of Neural Machine Translation (NMT) We find that the data scaling exponents are minimally impacted, suggesting that marginally worse architectures or training data can be compensated for by adding more data.
  arXiv  Detail & Related papers  (2022-02-04T06:53:49Z)
• Scaling Laws for Transfer [0.5432984841650929]
  We study scaling laws for transfer learning between distributions in an unsupervised, fine-tuning setting. We find that the effective data transferred is described well in the low data regime by a power-law of parameter count and fine-tuning dataset size.
  arXiv  Detail & Related papers  (2021-02-02T04:07:38Z)

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.