# LLMs Do Not See Age: Assessing Demographic Bias in Automated Systematic Review Synthesis

arXiv: https://arxiv.org/abs/2511.06000 · HTML: https://mlsanity.com/paper/2511.06000/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 37 references that are in the map.
- First posted on arXiv 8 Nov 2025.

## Builds on (most cited first, 16 of 16)

- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [BARTScore: Evaluating Generated Text as Text Generation](https://mlsanity.com/paper/2106.11520/) (2021)
- [Evaluating the Factual Consistency of Abstractive Text Summarization](https://mlsanity.com/paper/1910.12840/) (2019)
- [Mitigating Gender Bias in Natural Language Processing: Literature Review](https://mlsanity.com/paper/1906.08976/) (2019)
- [A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cycle](https://mlsanity.com/paper/1901.10002/) (2019)
- [Generating (Factual?) Narrative Summaries of RCTs: Experiments with Neural Multi-Document Summarization](https://mlsanity.com/paper/2008.11293/) (2020)
- [Re-Examining System-Level Correlations of Automatic Summarization Evaluation Metrics](https://mlsanity.com/paper/2204.10216/) (2022)
- [Investigating Subtler Biases in LLMs: Ageism, Beauty, Institutional, and Nationality Bias in Generative Models](https://mlsanity.com/paper/2309.08902/) (2023)
- [Unmasking and Quantifying Racial Bias of Large Language Models in Medical Report Generation](https://mlsanity.com/paper/2401.13867/) (2024)
- [Bias in News Summarization: Measures, Pitfalls and Corpora](https://mlsanity.com/paper/2309.08047/) (2023)
- [Evaluating Gender Bias of LLMs in Making Morality Judgements](https://mlsanity.com/paper/2410.09992/) (2024)
- [Examining Bias in Opinion Summarisation Through the Perspective of Opinion Diversity](https://mlsanity.com/paper/2306.04424/) (2023)
- [Bias in Opinion Summarisation from Pre-training to Adaptation: A Case Study in Political Bias](https://mlsanity.com/paper/2402.00322/) (2024)
- [Do Multi-Document Summarization Models Synthesize?](https://mlsanity.com/paper/2301.13844/) (2023)

## Built on by (most cited first, 5 of 5)

- [Faithful Summarisation under Disagreement via Belief-Level Aggregation](https://mlsanity.com/paper/2601.04889/) (2026)
- [Reinforcement-Learned Unequal Error Protection for Quantized Semantic Embeddings](https://mlsanity.com/paper/2601.00186/) (2026)
- [DIAL-SUMMER: A Structured Evaluation Framework of Hierarchical Errors in Dialogue Summaries](https://mlsanity.com/paper/2602.08149/) (2026)
- [Magic, Madness, Heaven, Sin: LLM Output Diversity is Everything, Everywhere, All at Once](https://mlsanity.com/paper/2604.01504/) (2026)
- [Pareto-Guided Teacher Alignment for Fair Personalized Text Generation](https://mlsanity.com/paper/2606.10126/) (2026)

## Abstract (arXiv)

Clinical interventions often hinge on age: medications and procedures safe for adults may be harmful to children or ineffective for older adults. However, as language models are increasingly integrated into biomedical evidence synthesis workflows, it remains uncertain whether these systems preserve such crucial demographic distinctions. To address this gap, we evaluate how well state-of-the-art language models retain age-related information when generating abstractive summaries of biomedical studies. We construct DemogSummary, a novel age-stratified dataset of systematic review primary studies, covering child, adult, and older adult populations. We evaluate three prominent summarisation-capable LLMs, Qwen (open-source), Longformer (open-source) and GPT-4.1 Nano (proprietary), using both standard metrics and a newly proposed Demographic Salience Score (DSS), which quantifies age-related entity retention and hallucination. Our results reveal systematic disparities across models and age groups: demographic fidelity is lowest for adult-focused summaries, and under-represented populations are more prone to hallucinations. These findings highlight the limitations of current LLMs in faithful and bias-free summarisation and point to the need for fairness-aware evaluation frameworks and summarisation pipelines in biomedical NLP.