# Do Prevalent Bias Metrics Capture Allocational Harms from LLMs?

arXiv: https://arxiv.org/abs/2408.01285 · HTML: https://mlsanity.com/paper/2408.01285/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 25 references that are in the map.
- First posted on arXiv 2 Aug 2024.

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

- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [TinyLlama: An Open-Source Small Language Model](https://mlsanity.com/paper/2401.02385/) (2024)
- [Algorithmic decision making and the cost of fairness](https://mlsanity.com/paper/1701.08230/) (2017)
- [Evaluating and Mitigating Discrimination in Language Model Decisions](https://mlsanity.com/paper/2312.03689/) (2023)
- [Stable LM 2 1.6B Technical Report](https://mlsanity.com/paper/2402.17834/) (2024)
- [Gender Bias in Multilingual Embeddings and Cross-Lingual Transfer](https://mlsanity.com/paper/2005.00699/) (2020)
- [Algorithmic Risk Assessments Can Alter Human Decision-Making Processes in High-Stakes Government Contexts](https://mlsanity.com/paper/2012.05370/) (2020)
- [Auditing the Use of Language Models to Guide Hiring Decisions](https://mlsanity.com/paper/2404.03086/) (2024)
- [Crowd, Lending, Machine, and Bias](https://mlsanity.com/paper/2008.04068/) (2020)

## Built on by (most cited first, 3 of 3)

- [Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure](https://mlsanity.com/paper/2603.16734/) (2026)
- [Responsible AI in NLP: GUS-Net Span-Level Bias Detection Dataset and Benchmark for Generalizations, Unfairness, and Stereotypes](https://mlsanity.com/paper/2410.08388/) (2024)
- [Need or Reputation? Region, National Income, and Academic Status in LLM Gatekeeping Decisions](https://mlsanity.com/paper/2608.05178/) (2026)

## Abstract (arXiv)

Allocational harms occur when resources or opportunities are unfairly withheld from specific groups. Many proposed bias measures ignore the discrepancy between predictions, which are what the proposed methods consider, and decisions that are made as a result of those predictions. Our work examines the reliability of current bias metrics in assessing allocational harms arising from predictions of large language models (LLMs). We evaluate their predictive validity and utility for model selection across ten LLMs and two allocation tasks. Our results reveal that commonly-used bias metrics based on average performance gap and distribution distance fail to reliably capture group disparities in allocation outcomes. Our work highlights the need to account for how model predictions are used in decisions, in particular in contexts where they are influenced by how limited resources are allocated.