# Measuring all the noises of LLM Evals

arXiv: https://arxiv.org/abs/2512.21326 · HTML: https://mlsanity.com/paper/2512.21326/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 25 references that are in the map.
- First posted on arXiv 24 Dec 2025.

## Builds on (most cited first, 12 of 12)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering](https://mlsanity.com/paper/2410.07095/) (2024)
- [tinyBenchmarks: evaluating LLMs with fewer examples](https://mlsanity.com/paper/2402.14992/) (2024)
- [CRUXEval: A Benchmark for Code Reasoning, Understanding and Execution](https://mlsanity.com/paper/2401.03065/) (2024)
- [Adding Error Bars to Evals: A Statistical Approach to Language Model Evaluations](https://mlsanity.com/paper/2411.00640/) (2024)
- [With Little Power Comes Great Responsibility](https://mlsanity.com/paper/2010.06595/) (2020)
- [SWE-Fixer: Training Open-Source LLMs for Effective and Efficient GitHub Issue Resolution](https://mlsanity.com/paper/2501.05040/) (2025)
- [Toward Training Superintelligent Software Agents through Self-Play SWE-RL](https://mlsanity.com/paper/2512.18552/) (2025)
- [Position: Don't Use the CLT in LLM Evals With Fewer Than a Few Hundred Datapoints](https://mlsanity.com/paper/2503.01747/) (2025)

## Built on by (most cited first, 7 of 7)

- [DeepSWE: Measuring Frontier Coding Agents on Original, Long-Horizon Engineering Tasks](https://mlsanity.com/paper/2607.07946/) (2026)
- [Brittlebench: Quantifying LLM robustness via prompt sensitivity](https://mlsanity.com/paper/2603.13285/) (2026)
- [Hidden Measurement Error in LLM Pipelines Distorts Annotation, Evaluation, and Benchmarking](https://mlsanity.com/paper/2604.11581/) (2026)
- [RevengeBench: Reverse Engineering Code-Space Policies from Behavioral Experiments](https://mlsanity.com/paper/2606.26094/) (2026)
- [Reinforcement Learning for Code Optimization](https://mlsanity.com/paper/2607.25970/) (2026)
- [When Pair Count Is Not the Sample Size: What All-Pairs Agent Comparisons Estimate](https://mlsanity.com/paper/2609.33012/) (2026)
- [Cortex: Measurement-Gated Evolution and Pruning of Coding-Agent Skills and Rules](https://mlsanity.com/paper/2610.14216/) (2026)

## Abstract (arXiv)

Applying well-established statistical methods effectively to LLM evals requires consideration of model correlations and prediction noises. Leading evals often ignore data noise, have low statistical power, or make unnecessary trade-offs. We define and measure three types of noise: 1) prediction noise from generating different answers on a given question, 2) data noise from sampling questions, and 3) their combined total noise following the law of total variance. To emphasize relative comparisons and gain statistical power, we propose all-pairs comparison, which applies paired comparison to all pairs of LLMs and measures all the noise components based on millions of question-level predictions across many evals and settings, revealing clear trends. First, most evals exhibit a characteristic and highly predictable total noise level across all model pairs. Second, paired prediction noise typically exceeds paired data noise, which means reducing prediction noise by averaging can significantly increase statistical power. By measuring all the noises together, we can assess eval results in context and make better empirical decisions.