Spend Less, Fit Better: Budget-Efficient Scaling Law Fitting via Active Experiment Selection

arXiv:2604.22753v1 Announce Type: new Abstract: Scaling laws are used to plan multi-million-dollar training runs, but fitting those laws can itself cost millions. In modern large-scale workflows, assembling a sufficiently informative set of pilot experiments is already a major budget-allocation problem rather than a routine preprocessing step. We formulate scaling-law fitting as budget-aware sequential experimental design: given a finite pool of runnable experiments with heterogeneous costs, choose which runs to execute so as to maximize extrapolation accuracy in a high-cost target region. We then propose an uncertainty-aware method for sequentially allocating experimental budget toward the runs most useful for target-region extrapolation. Across a diverse benchmark of scaling-law tasks, our method consistently outperforms classical design-based baselines, and often approaches the performance of fitting on the full experimental set while using only about 10% of the total training budget. Our code is available at https://github.com/PlanarG/active-sl.

Related

• Parcae: Scaling Laws For Stable Looped Language Models
• ProTrain: Efficient LLM Training via Memory-Aware Techniques
• MOONSHOT : A Framework for Multi-Objective Pruning of Vision and Large Language Models
• Low-rank Orthogonalization for Large-scale Matrix Optimization with Applications to Foundation Model Training
• Nexus: Same Pretraining Loss, Better Downstream Generalization via Common Minima

Source: arXiv cs.LG | 2026-04-27