A Steeper Slope Is Not Yet a New Law: What a Claimed Architectural Change to a Scaling Exponent Has to Survive, Why Studies of Looping, Growth, Routing and Optimizers Reach Opposite Verdicts, and the Floor, Recipe and Ladder Controls That Would Decide It

APA

Mahendrakar, P. (2026). A Steeper Slope Is Not Yet a New Law: What a Claimed Architectural Change to a Scaling Exponent Has to Survive, Why Studies of Looping, Growth, Routing and Optimizers Reach Opposite Verdicts, and the Floor, Recipe and Ladder Controls That Would Decide It [Preprint]. Life of Research. https://doi.org/10.5281/zenodo.23735381

Copy APA

BibTeX

@misc{lor1632026,
  title        = {A Steeper Slope Is Not Yet a New Law: What a Claimed Architectural Change to a Scaling Exponent Has to Survive, Why Studies of Looping, Growth, Routing and Optimizers Reach Opposite Verdicts, and the Floor, Recipe and Ladder Controls That Would Decide It},
  author       = {Pranay Mahendrakar},
  year         = {2026},
  publisher    = {Life of Research},
  doi          = {10.5281/zenodo.23735381},
  url          = {https://lifeofresearch.com/preprints/163},
  keywords     = {scaling laws, scaling exponents, looped transformers, model growth, compute-optimal training, irreducible loss, hyperparameter transfer, mixture of experts, pretraining},
  note         = {LOR-00163}
}

Copy BibTeX More on the record page