# Efficient Bayesian Inference from Noisy Pairwise Comparisons

arXiv: https://arxiv.org/abs/2510.09333 · HTML: https://mlsanity.com/paper/2510.09333/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 9 of its 38 references that are in the map.
- First posted on arXiv 10 Oct 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Human values and AI feedback alignment](https://mlsanity.com/alignment/human-values-and-ai-feedback-alignment/)

## Builds on (most cited first, 9 of 9)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [High-Fidelity Generative Image Compression](https://mlsanity.com/paper/2006.09965/) (2020)
- [Simple, Robust and Optimal Ranking from Pairwise Comparisons](https://mlsanity.com/paper/1512.08949/) (2015)
- [Double Thompson Sampling for Dueling Bandits](https://mlsanity.com/paper/1604.07101/) (2016)
- [Approximate Ranking from Pairwise Comparisons](https://mlsanity.com/paper/1801.01253/) (2018)
- [Good, Cheap, and Fast: Overfitted Image Compression with Wasserstein Distortion](https://mlsanity.com/paper/2412.00505/) (2024)
- [MOSPC: MOS Prediction Based on Pairwise Comparison](https://mlsanity.com/paper/2306.10493/) (2023)
- [Conditional Hallucinations for Image Compression](https://mlsanity.com/paper/2410.19493/) (2024)

## Built on by (most cited first, 2 of 2)

- [Calibrate, Don't Curate: Label-Efficient Estimation from Noisy LLM Judges](https://mlsanity.com/paper/2605.09702/) (2026)
- [LabelBuddy: An Open Source Music and Audio Language Annotation Tagging Tool Using AI Assistance](https://mlsanity.com/paper/2603.04293/) (2026)

## Abstract (arXiv)

Evaluating generative models is challenging because standard metrics often fail to reflect human preferences. Human evaluations are more reliable but costly and noisy, as participants vary in expertise, attention, and diligence. Pairwise comparisons improve consistency, yet aggregating them into overall quality scores requires careful modeling. Bradley-Terry-based methods update item scores from comparisons, but existing approaches either ignore rater variability or lack convergence guarantees, limiting robustness and interpretability. We introduce BBQ, a Bayesian Bradley-Terry variant that explicitly models rater quality, downweighting or removing unreliable participants, and provides guaranteed monotonic likelihood convergence through an Expectation-Maximization algorithm. Empirical results show that BBQ provides efficient inference, well-calibrated uncertainty estimates, and more robust, interpretable rankings compared to baseline Bradley-Terry models, even with noisy or crowdsourced raters. This framework enables more reliable and cost-effective human evaluation of generative models.