English

FLOP-Efficient Training: Early Stopping Based on Test-Time Compute Awareness

Computation and Language 2026-01-06 v1 Machine Learning

Abstract

Scaling training compute, measured in FLOPs, has long been shown to improve the accuracy of large language models, yet training remains resource-intensive. Prior work shows that increasing test-time compute (TTC)-for example through iterative sampling-can allow smaller models to rival or surpass much larger ones at lower overall cost. We introduce TTC-aware training, where an intermediate checkpoint and a corresponding TTC configuration can together match or exceed the accuracy of a fully trained model while requiring substantially fewer training FLOPs. Building on this insight, we propose an early stopping algorithm that jointly selects a checkpoint and TTC configuration to minimize training compute without sacrificing accuracy. To make this practical, we develop an efficient TTC evaluation method that avoids exhaustive search, and we formalize a break-even bound that identifies when increased inference compute compensates for reduced training compute. Experiments demonstrate up to 92\% reductions in training FLOPs while maintaining and sometimes remarkably improving accuracy. These results highlight a new perspective for balancing training and inference compute in model development, enabling faster deployment cycles and more frequent model refreshes. Codes will be publicly released.

Keywords

Cite

@article{arxiv.2601.01332,
  title  = {FLOP-Efficient Training: Early Stopping Based on Test-Time Compute Awareness},
  author = {Hossam Amer and Maryam Dialameh and Hossein Rajabzadeh and Walid Ahmed and Weiwei Zhang and Yang Liu},
  journal= {arXiv preprint arXiv:2601.01332},
  year   = {2026}
}
R2 v1 2026-07-01T08:49:35.713Z