# Robust AI Evaluation through Maximal Lotteries

arXiv: https://arxiv.org/abs/2602.21297 · HTML: https://mlsanity.com/paper/2602.21297/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 50 references that are in the map.
- First posted on arXiv 24 Feb 2026.

## Builds on (most cited first, 30 of 32)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [AgentBench: Evaluating LLMs as Agents](https://mlsanity.com/paper/2308.03688/) (2023)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators](https://mlsanity.com/paper/2404.04475/) (2024)
- [Learning Models with Uniform Performance via Distributionally Robust Optimization](https://mlsanity.com/paper/1810.08750/) (2018)
- [Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators](https://mlsanity.com/paper/2403.16950/) (2024)
- [A Minimaximalist Approach to Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2401.04056/) (2024)
- [The Benchmark Lottery](https://mlsanity.com/paper/2107.07002/) (2021)
- [Contextual Dueling Bandits](https://mlsanity.com/paper/1502.06362/) (2015)
- [Social Choice Should Guide AI Alignment in Dealing with Diverse Human Feedback](https://mlsanity.com/paper/2404.10271/) (2024)
- [Open-LLM-Leaderboard: From Multi-choice to Open-style Questions for LLMs Evaluation, Benchmark, and Arena](https://mlsanity.com/paper/2406.07545/) (2024)
- [Elo Uncovered: Robustness and Best Practices in Language Model Evaluation](https://mlsanity.com/paper/2311.17295/) (2023)
- [Axioms for AI Alignment from Human Feedback](https://mlsanity.com/paper/2405.14758/) (2024)
- [Diverging Preferences: When do Annotators Disagree and do Models Know?](https://mlsanity.com/paper/2410.14632/) (2024)
- [Investigating Non-Transitivity in LLM-as-a-Judge](https://mlsanity.com/paper/2502.14074/) (2025)
- [Search Arena: Analyzing Search-Augmented LLMs](https://mlsanity.com/paper/2506.05334/) (2025)
- [Examining the robustness of LLM evaluation to the distributional assumptions of benchmarks](https://mlsanity.com/paper/2404.16966/) (2024)
- [Inherent Trade-Offs between Diversity and Stability in Multi-Task Benchmarks](https://mlsanity.com/paper/2405.01719/) (2024)
- [Mapping Social Choice Theory to RLHF](https://mlsanity.com/paper/2404.13038/) (2024)
- [Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment](https://mlsanity.com/paper/2410.02197/) (2024)
- [Pairwise Calibrated Rewards for Pluralistic Alignment](https://mlsanity.com/paper/2506.06298/) (2025)
- [A Statistical Framework for Ranking LLM-Based Chatbots](https://mlsanity.com/paper/2412.18407/) (2024)
- [am-ELO: A Stable Framework for Arena-based LLM Evaluation](https://mlsanity.com/paper/2505.03475/) (2025)
- [Robust Stackelberg Equilibria in Extensive-Form Games and Extension to Limited Lookahead](https://mlsanity.com/paper/1711.08080/) (2017)

## Built on by (most cited first, 1 of 1)

- [AI Alignment From Social Choice Perspectives](https://mlsanity.com/paper/2606.21550/) (2026)

## Abstract (arXiv)

The standard way to evaluate language models on subjective tasks is through pairwise comparisons: an annotator chooses the "better" of two responses to a prompt. Leaderboards aggregate these comparisons into a single Bradley-Terry (BT) ranking, forcing heterogeneous preferences into a total order and violating basic social-choice desiderata. In contrast, social choice theory provides an alternative approach called maximal lotteries, which aggregates pairwise preferences without imposing any assumptions on their structure. However, we show that maximal lotteries are highly sensitive to preference heterogeneity and can favor models that severely underperform on specific tasks or user subpopulations. We introduce robust lotteries that optimize worst-case performance under plausible shifts in the preference data. On large-scale preference datasets, robust lotteries provide more reliable win rate guarantees across the annotator distribution and recover a stable set of top-performing models. By moving from rankings to pluralistic sets of winners, robust lotteries offer a principled step toward an ecosystem of complementary AI systems that serve the full spectrum of human preferences.