arXiv Machine Learning By Gabriel Sargent, Will Wei Sun, Zhengwu Zhang, Yufeng Liu

Prediction-Powered LLM Ranking with Adaptively Collected Human Feedback

Read the original on arXiv Machine Learning →

The paper introduces Prediction-Powered Ranking (PP‑Rank), a framework that blends large language model (LLM) judge predictions with limited human feedback to rank LLMs more precisely while maintaining valid inference under adaptive sampling. PP‑Rank uses data‑driven prediction weights to minimize asymptotic covariance, establishes asymptotic normality for general evaluation targets, and provides a rule for reducing exploration that balances cost and variance. Experiments on UltraFeedback‑Extended and synthetic data show that PP‑Rank improves score estimation, narrows confidence intervals, certifies rankings, and can cut evaluation costs without losing precision.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv Statistics ML
Sep 4

Low Rank for Rank: Uncertainty-Aware Task-Specific LLM Ranking under Sparse Pairwise Comparisons

The paper introduces a low‑rank framework for ranking large language models (LLMs) on task‑specific benchmarks using sparse pairwise comparisons. By modeling the task‑by‑model ability matrix as low rank, the method shares information across related tasks while preserving task‑specific differences, and it provides uncertainty‑aware ranking through debiased estimators and simultaneous confidence sets. Experiments on synthetic data and the Chatbot Arena benchmark demonstrate improved sample efficiency and tighter, better‑calibrated ranking certificates, especially in the sparse comparison regime typical of real LLM evaluations.

By Jiachun Li, David Simchi-Levi, Will Wei Sun