# Variance reduction in output from generative AI

arXiv: https://arxiv.org/abs/2503.01033 · HTML: https://mlsanity.com/paper/2503.01033/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 48 references that are in the map.
- First posted on arXiv 2 Mar 2025.

## Builds on (most cited first, 25 of 25)

- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [A Survey of Large Language Models](https://mlsanity.com/paper/2303.18223/) (2023)
- [Diffusion Models in Vision: A Survey](https://mlsanity.com/paper/2209.04747/) (2022)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [Persistent Anti-Muslim Bias in Large Language Models](https://mlsanity.com/paper/2101.05783/) (2021)
- [Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers](https://mlsanity.com/paper/2409.04109/) (2024)
- [ToolQA: A Dataset for LLM Question Answering with External Tools](https://mlsanity.com/paper/2306.13304/) (2023)
- [Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale](https://mlsanity.com/paper/2211.03759/) (2022)
- [Evaluating Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2310.19736/) (2023)
- ["Kelly is a Warm Person, Joseph is a Role Model": Gender Biases in LLM-Generated Reference Letters](https://mlsanity.com/paper/2310.09219/) (2023)
- [Gender bias and stereotypes in Large Language Models](https://mlsanity.com/paper/2308.14921/) (2023)
- [Should ChatGPT be Biased? Challenges and Risks of Bias in Large Language Models](https://mlsanity.com/paper/2304.03738/) (2023)
- [On the Use of ArXiv as a Dataset](https://mlsanity.com/paper/1905.00075/) (2019)
- [Better to Ask in English: Cross-Lingual Evaluation of Large Language Models for Healthcare Queries](https://mlsanity.com/paper/2310.13132/) (2023)
- [AI and the Problem of Knowledge Collapse](https://mlsanity.com/paper/2404.03502/) (2024)
- [LLMRec: Benchmarking Large Language Models on Recommendation Task](https://mlsanity.com/paper/2308.12241/) (2023)
- [BertaQA: How Much Do Language Models Know About Local Culture?](https://mlsanity.com/paper/2406.07302/) (2024)
- [Bias in Generative AI](https://mlsanity.com/paper/2403.02726/) (2024)
- [Improving Cross-lingual Information Retrieval on Low-Resource Languages via Optimal Transport Distillation](https://mlsanity.com/paper/2301.12566/) (2023)
- [Bias and Volatility: A Statistical Framework for Evaluating Large Language Model's Stereotypes and the Associated Generation Inconsistency](https://mlsanity.com/paper/2402.15481/) (2024)
- [The LLM Effect: Are Humans Truly Using LLMs, or Are They Being Influenced By Them Instead?](https://mlsanity.com/paper/2410.04699/) (2024)
- [Generative Discrimination: What Happens When Generative AI Exhibits Bias, and What Can Be Done About It](https://mlsanity.com/paper/2407.10329/) (2024)
- [Measuring Human Contribution in AI-Assisted Content Generation](https://mlsanity.com/paper/2408.14792/) (2024)

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

- [The Homogenization Problem in LLMs: Towards Meaningful Diversity in AI Safety](https://mlsanity.com/paper/2601.06116/) (2026)
- [Inducing Sustained Creativity and Diversity in Large Language Models](https://mlsanity.com/paper/2603.19519/) (2026)
- [The Variance Paradox: How AI Reduces Diversity but Increases Novelty](https://mlsanity.com/paper/2508.19264/) (2025)
- [A Systematic Evaluation of Large Language Models for PTSD Severity Estimation: The Role of Contextual Knowledge and Modeling Strategies](https://mlsanity.com/paper/2602.06015/) (2026)
- [The average-farmer illusion in language-model simulations of agricultural decisions](https://mlsanity.com/paper/2609.15038/) (2026)
- [Fine-Tuning Fixes Mode Collapse and Over-Dispersion in LLMs](https://mlsanity.com/paper/2609.16454/) (2026)

## Abstract (arXiv)

Generative AI models, such as ChatGPT, will increasingly replace humans in producing output for a variety of important tasks. While much prior work has mostly focused on the improvement in the average performance of generative AI models relative to humans' performance, much less attention has been paid to the significant reduction of variance in output produced by generative AI models. In this Perspective, we demonstrate that generative AI models are inherently prone to the phenomenon of "regression toward the mean" whereby variance in output tends to shrink relative to that in real-world distributions. We discuss potential social implications of this phenomenon across three levels-societal, group, and individual-and two dimensions-material and non-material. Finally, we discuss interventions to mitigate negative effects, considering the roles of both service providers and users. Overall, this Perspective aims to raise awareness of the importance of output variance in generative AI and to foster collaborative efforts to meet the challenges posed by the reduction of variance in output generated by AI models.