Dropping Just a Handful of Preferences Can Change Top Large Language Model Rankings

arXiv:2508.11847 · stat.ML, cs.LG · Submitted 2025-08-16 · Read on arXiv

stat.ML, cs.LG

Submitted: 2025-08-16

Updated: 2026-07-09

Terminology

• Robustness — 2× in this paper · explained in 6 episodes
• LLM-as-a-Judge — 1× in this paper · explained in 4 episodes

Related papers

• Behavior of prediction performance metrics with rare events
• Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
• A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
• Scalable Posterior Uncertainty for Flexible Density-Based Clustering
• Efficient quadratic entropy with distance sketches
• Embedding-Bias in Conditional Independence Testing