AI Paper Atlas

Research map / Large language models / AI and society

AI Cartography: Mapping the Latent Landscape of AI Benchmark Ecosystems

Michael Hardy, Anka Reuel, Lijin Zhang, Jodi M. Casabianca, Sang T. Truong, Yash Satish Dave, Hansol Lee, Benjamin W. Domingue, Sanmi Koyejo

ICML 2026 · Large language models › AI and society · approach: Social-science view of AI evaluation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fit a bifactor structure (general plus benchmark-specific factors) instead of independent or unidimensional structures (ablation)CFA of Open LLM Leaderboard items · held-out AUCAUC: 0.90 -> 0.95Improved
allow correlated specific factors on top of the bifactor (correlated bifactor) (ablation)bifactor CFA · held-out AUCAUC: 0.95 -> 0.77Hurt
allow correlated specific factors on top of the bifactor (correlated bifactor) (ablation)bifactor CFA · RMSEARMSEA: 0.05 -> 0.04Improved
randomize item-to-benchmark assignment (permutation control) (ablation)flexible SEM structures · item-level CFAfitNo clear effect
add size random slopes and control for model size in the variance decomposition (ablation)G-study · Open LLM Leaderboardcontributor variance share: 0.086 -> 0.078No clear effect
estimate scaling slope on manifest scores instead of latent general factorscaling law · Open LLM Leaderboardslope reliability: 0.97 -> 0.53Hurt
control for deployment and contributor effects in latent regression of g on size (ablation)bifactor latent scores · general factor scalingsize slopeImproved
increase model size (log10 params), after controlsMMLU-Pro specific factor · latent regressionslope beta: beta 0.52Improved
increase model size (log10 params), after controlsBBH and GPQA specific factors · latent regressionslope beta: BBH -0.05, GPQA 0.07No clear effect
increase model size (log10 params), after controlsMuSR specific factor · latent regressionslope beta: beta 0.04No clear effect
apply chat template / instruction fine-tuningIFEval specific factor · latent regressioneffect gamma: gamma 1.82Improved
apply chat template / instruction fine-tuningMuSR specific factor · latent regressioneffect gamma: gamma -1.03Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy