Diagnostic of physics literacy in frontier LLMs. Evaluates induction, formulation, and prediction inside unfamiliar physics frameworks via dual-judge inter-rater reliability with human-audit resolution.
CTA-Bench: research benchmark and toolkit for studying how well systems turn problem descriptions and reference code into Lean 4 proof obligations, and how faithful those obligations are to the intended algorithm.
Code, data, and paper for an evaluation-protocol-adjusted residual audit of public LLM benchmark scores — predicting score variation from public metadata and using the residuals as a review queue.
MCDM using TOPSIS: A technique for multi-criteria decision making, utilizing the TOPSIS method to evaluate alternatives based on multiple criteria.
The instrument makes the winner: reconciling two contradictory 2026 clinical-AI evaluations (Real-POCQi vs Nature Medicine). A 2x2 instrument x rater decomposition shows the evaluation instrument (pairwise vs rubric), not the model, drives the disagreement.
Runtime-verifiable audit harness that falsifies coding-agent intervention claims — preregistered, oracle-checked, reports null.
Add a description, image, and links to the evaluation-methodology topic page so that developers can more easily learn about it.
To associate your repository with the evaluation-methodology topic, visit your repo's landing page and select "manage topics."