Preregister for the next LLM
LLM experiments are easy to p-hack, that is, tune after results are visible, by changing the model, prompt, etc. until a preferred result appears. This initiative instead exploits new model releases as a holdout set. It asks to preregister the full procedure and then run it on the next eligible model, thereby letting researchers provide a certificate against p-hacking the model that is used.
Maria Thomas, Kristina Gligorić, Nihar B. Shah (2026). arXiv:2606.27687.
Effectiveness of the protocol
Complete the AsPredicted template
Use the initiative’s AsPredicted template at AsPredicted.org to create a timestamped preregistration of the configuration and eligible models.
Select “No.” The chosen future model must still be unavailable, so confirmatory outputs cannot yet exist.
Record the predicted result, any directional prediction, and the threshold for support. Define eligible families and variants, plus a selection rule that an independent reader can apply to future releases.
Provide a deterministic formula that converts acceptable model responses into the study outcome. Specify the required syntax, label set, parsing procedure, and normalization.
Record every model-use setting: the full prompt, decoding controls, token limit, and whether execution is batched or item-by-item.
Name the planned statistical procedures and the decision rules used to interpret them.
List every invalid-output case and the deterministic response to it. Set a model-level failure threshold, the switch-to-next-model rule, and any additional validity checks.
Give the item count and identify the dataset and version precisely enough to reconstruct the analyzed sample.
Precommit to any follow-up models and to rules for model failure, newly introduced model families, and discontinued families.
Compare a preregistration and paper
The audit reports whether the paper uses the first eligible model released after preregistration, whether the reported configuration matches the preregistration, and whether the preregistration follows the template.
The preregistration PDF, paper PDF, and API key are sent directly to the selected model provider for this request. This page does not store them.