What Counts as a Control? A Scaffold Effect Is a Contrast Conditional on What Was Held Fixed, the 2026 Controlled Studies Each Held Something Different, and Only a Declared Spend Interval Makes a Harness Claim Refutable

APA

Mahendrakar, P. (2026). What Counts as a Control? A Scaffold Effect Is a Contrast Conditional on What Was Held Fixed, the 2026 Controlled Studies Each Held Something Different, and Only a Declared Spend Interval Makes a Harness Claim Refutable [Journal article]. Life of Research. https://doi.org/10.5281/zenodo.23306459

Copy APA

BibTeX

@misc{lor1032026,
  title        = {What Counts as a Control? A Scaffold Effect Is a Contrast Conditional on What Was Held Fixed, the 2026 Controlled Studies Each Held Something Different, and Only a Declared Spend Interval Makes a Harness Claim Refutable},
  author       = {Pranay Mahendrakar},
  year         = {2026},
  publisher    = {Life of Research},
  doi          = {10.5281/zenodo.23306459},
  url          = {https://lifeofresearch.com/articles/103},
  keywords     = {agent evaluation, agent harness, scaffolding, compute-matched baselines, cost-controlled evaluation, experimental design, benchmark validity, LLM agents},
  note         = {LOR-00103}
}

Copy BibTeX More on the record page