# Selection Bias Induced Spurious Correlations in Large Language Models

arXiv: https://arxiv.org/abs/2207.08982 · HTML: https://mlsanity.com/paper/2207.08982/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 15 references that are in the map.
- First posted on arXiv 18 Jul 2022.

## Builds on (most cited first, 6 of 6)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Underspecification Presents Challenges for Credibility in Modern Machine Learning](https://mlsanity.com/paper/2011.03395/) (2020)
- [Gender Bias in Coreference Resolution](https://mlsanity.com/paper/1804.09301/) (2018)
- [Measuring and Reducing Gendered Correlations in Pre-trained Models](https://mlsanity.com/paper/2010.06032/) (2020)
- [Neural Text Generation from Structured Data with Application to the Biography Domain](https://mlsanity.com/paper/1603.07771/) (2016)

## Built on by (most cited first, 9 of 9)

- [Causal Parrots: Large Language Models May Talk Causality But Are Not Causal](https://mlsanity.com/paper/2308.13067/) (2023)
- [Beyond Reward Hacking: Causal Rewards for Large Language Model Alignment](https://mlsanity.com/paper/2501.09620/) (2025)
- [MM-SpuBench: Towards Better Understanding of Spurious Biases in Multimodal LLMs](https://mlsanity.com/paper/2406.17126/) (2024)
- [KALM: Knowledge-Aware Integration of Local, Document, and Global Contexts for Long Document Understanding](https://mlsanity.com/paper/2210.04105/) (2022)
- [Debiasing Reward Models by Representation Learning with Guarantees](https://mlsanity.com/paper/2510.23751/) (2025)
- [CoBA: Counterbias Text Augmentation for Mitigating Various Spurious Correlations via Semantic Triples](https://mlsanity.com/paper/2508.21083/) (2025)
- [Selection Collider Bias in Large Language Models](https://mlsanity.com/paper/2208.10063/) (2022)
- [Evaluation of Large Language Models: STEM education and Gender Stereotypes](https://mlsanity.com/paper/2406.10133/) (2024)
- [Recover Experimental Data with Selection Bias using Counterfactual Logic](https://mlsanity.com/paper/2506.00335/) (2025)

## Abstract (arXiv)

In this work we show how large language models (LLMs) can learn statistical dependencies between otherwise unconditionally independent variables due to dataset selection bias. To demonstrate the effect, we developed a masked gender task that can be applied to BERT-family models to reveal spurious correlations between predicted gender pronouns and a variety of seemingly gender-neutral variables like date and location, on pre-trained (unmodified) BERT and RoBERTa large models. Finally, we provide an online demo, inviting readers to experiment further.