# Evaluating how LLM annotations represent diverse views on contentious topics

arXiv: https://arxiv.org/abs/2503.23243 · HTML: https://mlsanity.com/paper/2503.23243/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 52 references that are in the map.
- First posted on arXiv 29 Mar 2025.

## Builds on (most cited first, 30 of 31)

- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Semantics derived automatically from language corpora contain human-like biases](https://mlsanity.com/paper/1608.07187/) (2016)
- [ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks](https://mlsanity.com/paper/2303.15056/) (2023)
- [The Platonic Representation Hypothesis](https://mlsanity.com/paper/2405.07987/) (2024)
- [Dealing with Disagreements: Looking Beyond the Majority Vote in Subjective Annotations](https://mlsanity.com/paper/2110.05719/) (2021)
- [Word Embeddings Quantify 100 Years of Gender and Ethnic Stereotypes](https://mlsanity.com/paper/1711.08412/) (2017)
- [LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks](https://mlsanity.com/paper/2406.18403/) (2024)
- [Ex Machina: Personal Attacks Seen at Scale](https://mlsanity.com/paper/1610.08914/) (2016)
- [Large Language Models for Data Annotation and Synthesis: A Survey](https://mlsanity.com/paper/2402.13446/) (2024)
- [Gender Bias in Contextualized Word Embeddings](https://mlsanity.com/paper/1904.03310/) (2019)
- [Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMs](https://mlsanity.com/paper/2311.04892/) (2023)
- [Gender bias and stereotypes in Large Language Models](https://mlsanity.com/paper/2308.14921/) (2023)
- [Quantifying the Persona Effect in LLM Simulations](https://mlsanity.com/paper/2402.10811/) (2024)
- [On Releasing Annotator-Level Labels and Information in Datasets](https://mlsanity.com/paper/2110.05699/) (2021)
- [Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselves](https://mlsanity.com/paper/2311.04205/) (2023)
- [ChatGPT-4 Outperforms Experts and Crowd Workers in Annotating Political Twitter Messages with Zero-Shot Learning](https://mlsanity.com/paper/2304.06588/) (2023)
- [Can LLM be a Personalized Judge?](https://mlsanity.com/paper/2406.11657/) (2024)
- [NLPositionality: Characterizing Design Biases of Datasets and Models](https://mlsanity.com/paper/2306.01943/) (2023)
- [CoAnnotating: Uncertainty-Guided Work Allocation between Human and Large Language Models for Data Annotation](https://mlsanity.com/paper/2310.15638/) (2023)
- [When "A Helpful Assistant" Is Not Really Helpful: Personas in System Prompts Do Not Improve Performances of Large Language Models](https://mlsanity.com/paper/2311.10054/) (2023)
- [Everyone's Voice Matters: Quantifying Annotation Disagreement Using Demographic Information](https://mlsanity.com/paper/2301.05036/) (2023)
- [When Do Annotator Demographics Matter? Measuring the Influence of Annotator Demographics with the POPQUORN Dataset](https://mlsanity.com/paper/2306.06826/) (2023)
- [The Ecological Fallacy in Annotation: Modelling Human Label Variation goes beyond Sociodemographics](https://mlsanity.com/paper/2306.11559/) (2023)
- ["HOT" ChatGPT: The promise of ChatGPT in detecting and discriminating hateful, offensive, and toxic comments on social media](https://mlsanity.com/paper/2304.10619/) (2023)
- [The Unequal Opportunities of Large Language Models: Revealing Demographic Bias through Job Recommendations](https://mlsanity.com/paper/2308.02053/) (2023)
- [Sociodemographic Prompting is Not Yet an Effective Approach for Simulating Subjective Judgments with LLMs](https://mlsanity.com/paper/2311.09730/) (2023)
- [Unveiling Gender Bias in Terms of Profession Across LLMs: Analyzing and Addressing Sociological Implications](https://mlsanity.com/paper/2307.09162/) (2023)
- [Ruddit: Norms of Offensiveness for English Reddit Comments](https://mlsanity.com/paper/2106.05664/) (2021)
- [Prompt Design Matters for Computational Social Science Tasks but in Unpredictable Ways](https://mlsanity.com/paper/2406.11980/) (2024)
- [What Do Llamas Really Think? Revealing Preference Biases in Language Model Representations](https://mlsanity.com/paper/2311.18812/) (2023)

## Built on by (most cited first, 6 of 6)

- [Modeling Annotator Disagreement with Demographic-Aware Experts and Synthetic Perspectives](https://mlsanity.com/paper/2508.02853/) (2025)
- [Algorithmic Fairness in NLP: Persona-Infused LLMs for Human-Centric Hate Speech Detection](https://mlsanity.com/paper/2510.19331/) (2025)
- [When Persona Attributes Improve Population Alignment in Large Language Models](https://mlsanity.com/paper/2609.02526/) (2026)
- [Learning Who Disagrees: Demographic Importance Weighting for Modeling Annotator Distributions with DiADEM](https://mlsanity.com/paper/2604.08425/) (2026)
- [From Self to Other: Evaluating Demographic Perspective-Taking in LLM Hate Speech Annotation](https://mlsanity.com/paper/2606.06266/) (2026)
- [Demographic Prompting at Scale: When More Attributes Hurt LLM--Human Agreement](https://mlsanity.com/paper/2607.10590/) (2026)

## Abstract (arXiv)

Researchers have proposed the use of generative large language models (LLMs) to label data for research and applied settings. This literature emphasizes the improved performance of these models relative to other natural language models, noting that generative LLMs typically outperform other models and even humans across several metrics. Previous literature has examined bias across many applications and contexts, but less work has focused specifically on bias in generative LLMs' responses to subjective annotation tasks. This bias could result in labels applied by LLMs that disproportionately align with majority groups over a more diverse set of viewpoints. In this paper, we evaluate how LLMs represent diverse viewpoints on these contentious tasks. Across four annotation tasks on four datasets, we show that LLMs do not show systematic substantial disagreement with annotators on the basis of demographics. Rather, we find that multiple LLMs tend to be biased in the same directions on the same demographic categories within the same datasets. Moreover, the disagreement between human annotators on the labeling task -- a measure of item difficulty -- is far more predictive of LLM agreement with human annotators. We conclude with a discussion of the implications for researchers and practitioners using LLMs for automated data annotation tasks. Specifically, we emphasize that fairness evaluations must be contextual, model choice alone will not solve potential issues of bias, and item difficulty must be integrated into bias assessments.