mlsanityData to 14 Oct 2026

Biases in Large Language Model-Elicited Text: A Case Study in Natural Language Inference

arXiv 2503.05047 ↗First posted 6 Mar 2025cs.CLcited by 1

Not on the map. Why?

Abstract (arXiv)

We test whether NLP datasets created with Large Language Models (LLMs) contain annotation artifacts and social biases like NLP datasets elicited from crowd-source workers. We recreate a portion of the Stanford Natural Language Inference corpus using GPT-4, Llama-2 70b for Chat, and Mistral 7b Instruct. We train hypothesis-only classifiers to determine whether LLM-elicited NLI datasets contain annotation artifacts. Next, we use pointwise mutual information to identify the words in each dataset that are associated with gender, race, and age-related terms. On our LLM-generated NLI datasets, fine-tuned BERT hypothesis-only classifiers achieve between 86-96% accuracy. Our analyses further characterize the annotation artifacts and stereotypical biases in LLM-generated datasets.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 34 of its 62 references that are in the map. First posted on arXiv 6 Mar 2025.

Builds on

34 of 62 references.

References · 34

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  3. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
  4. 2022BLOOM: A 176B-Parameter Open-Access Multilingual Language Modelcited by 1,984
  5. 2015A large annotated corpus for learning natural language inferencecited by 1,847
  6. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,729
  7. 2022In-context Learning and Induction Headscited by 964
  8. 2023ChatGPT Outperforms Crowd-Workers for Text-Annotation Taskscited by 870
  9. 2018Annotation Artifacts in Natural Language Inference Datacited by 671
  10. 2023Bias and Fairness in Large Language Models: A Surveycited by 639
  11. 2024OLMo: Accelerating the Science of Language Modelscited by 613
  12. 2023The Falcon Series of Open Language Modelscited by 535
22 more
  1. 2023Benchmarking Large Language Models for News Summarizationcited by 445
  2. 2016A Corpus and Evaluation Framework for Deeper Understanding of Commonsense Storiescited by 417
  3. 2022Unnatural Instructions: Tuning Language Models with (Almost) No Human Laborcited by 397
  4. 2018Hypothesis Only Baselines in Natural Language Inferencecited by 295
  5. 2021Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Modelscited by 222
  6. 2022ZeroGen: Efficient Zero-shot Learning via Dataset Generationcited by 211
  7. 2021Societal Biases in Language Generation: Progress and Challengescited by 202
  8. 2022WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creationcited by 202
  9. 2021Reframing Human-AI Collaboration for Generating Free-Text Explanationscited by 125
  10. 2023Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventionscited by 125
  11. 2023Generative Language Models Exhibit Social Identity Biasescited by 111
  12. 2023Artificial Artificial Artificial Intelligence: Crowd Workers Widely Use Large Language Models for Text Production Taskscited by 103
  13. 2018How Much Reading Does Reading Comprehension Require? A Critical Investigation of Popular Benchmarkscited by 88
  14. 2024Evaluating Large Language Model Biases in Persona-Steered Generationcited by 78
  15. 2023Automated Annotation with Generative AI Requires Validationcited by 76
  16. 2022Rethinking the Role of Scale for In-Context Learning: An Interpretability-based Case Study at 66 Billion Scalecited by 64
  17. 2018Performance Impact Caused by Hidden Bias of Training Data for Recognizing Textual Entailmentcited by 59
  18. 2023Bias Against 93 Stigmatized Groups in Masked Language Models and Downstream Sentiment Classification Taskscited by 51
  19. 2017The Effect of Different Writing Tasks on Linguistic Style: A Case Study of the ROC Story Cloze Taskcited by 39
  20. 2023The Parrot Dilemma: Human-Labeled vs. LLM-augmented Data in Classification Taskscited by 35
  21. 2023Large Language Models as Annotators: Enhancing Generalization of NLP Models at Minimal Costcited by 33
  22. 2024Breaking Bias, Building Bridges: Evaluation and Mitigation of Social Biases in LLMs via Contact Hypothesiscited by 20

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Semantic Abstraction for Natural Language Inference: a Methodological Framework for Discovering and Compensating Semantic Knowledge and Reasoning Gaps in Large Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Hypothesis-only Biases in Large Language Model-Elicited Natural Language Inference
  2. 2024Interpreting Bias in Large Language Models: A Feature-Based Approach
  3. 2025Dissecting Bias in LLMs: A Mechanistic Interpretability Perspective
  4. 2024BEADs: Bias Evaluation Across Domains
  5. 2024Decoding Biases: Automated Methods and LLM Judges for Gender Bias Detection in Language Models
  6. 2024Towards Resource Efficient and Interpretable Bias Mitigation in Large Language Models
  7. 2025Unsupervised Concept Vector Extraction for Bias Control in LLMs
  8. 2024Evaluating Nuanced Bias in Large Language Model Free Response Answers