mlsanityData to 14 Oct 2026

Spurious Correlations and Beyond: Understanding and Mitigating Shortcut Learning in SDOH Extraction with Large Language Models

arXiv 2506.00134 ↗First posted 30 May 2025cs.CL cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

Social determinants of health (SDOH) extraction from clinical text is critical for downstream healthcare analytics. Although large language models (LLMs) have shown promise, they may rely on superficial cues leading to spurious predictions. Using the MIMIC portion of the SHAC (Social History Annotation Corpus) dataset and focusing on drug status extraction as a case study, we demonstrate that mentions of alcohol or smoking can falsely induce models to predict current/past drug use where none is present, while also uncovering concerning gender disparities in model performance. We further evaluate mitigation strategies - such as prompt engineering and chain-of-thought reasoning - to reduce these false positives, providing insights into enhancing LLM reliability in health domains.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 16 of its 40 references that are in the map. First posted on arXiv 30 May 2025.

Builds on

16 of 40 references.

References · 16

  1. 2020Shortcut Learning in Deep Neural Networkscited by 1,969
  2. 2022Large Language Models Encode Clinical Knowledgecited by 1,830
  3. 2020Beyond Accuracy: Behavioral Testing of NLP models with CheckListcited by 1,006
  4. 2019Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inferencecited by 876
  5. 2017Adversarial Examples for Evaluating Reading Comprehension Systemscited by 571
  6. 2017Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraintscited by 420
  7. 2020An Empirical Study on Robustness to Spurious Correlations using Pre-trained Language Modelscited by 134
  8. 2023DeID-GPT: Zero-shot Medical Text De-Identification by GPT-4cited by 130
  9. 2023Large Language Model Is Not a Good Few-shot Information Extractor, but a Good Reranker for Hard Samples!cited by 119
  10. 2024Med42-v2: A Suite of Clinical LLMscited by 90
  11. 2023Large Language Models Can be Lazy Learners: Analyze Shortcuts in In-Context Learningcited by 69
  12. 2022Detecting Shortcut Learning for Fair Medical AI using Shortcut Testingcited by 52
4 more
  1. 2020Annotating Social Determinants of Health Using Active Learning, and Characterizing Determinants Using Neural Event Extractioncited by 32
  2. 2020Deep Learning Applied to Chest X-Rays: Exploiting and Preventing Shortcutscited by 29
  3. 2022DICE: Data-Efficient Clinical Event Extraction with Generative Modelscited by 20
  4. 2021A Study of Social and Behavioral Determinants of Health in Lung Cancer Patients Using Transformers-based Natural Language Processing Modelscited by 11

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Clinical Note Bloat Reduction for Efficient LLM Usecited by 2
  2. 2026Benchmarking Fairness in Spiking Neural Networks: Data Bias, Spurious Features, and Hardware Effectscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Social Determinants of Health Prediction for ICD-9 Code with Reasoning Models
  2. 2025A Multi-Stage Large Language Model Framework for Extracting Suicide-Related Social Determinants of Health
  3. 2024Synth-SBDH: A Synthetic Dataset of Social and Behavioral Determinants of Health for Clinical Text
  4. 2026Using reasoning LLMs to extract SDOH events from clinical notes
  5. 2026Structured Insight from Unstructured Data: Large Language Models for SDOH-Driven Diabetes Risk Prediction
  6. 2023Distilling Large Language Models for Biomedical Knowledge Extraction: A Case Study on Adverse Drug Events
  7. 2025Interpretable Graph-Language Modeling for Detecting Youth Illicit Drug Use
  8. 2026Large language models can disambiguate opioid slang on social media