mlsanityData to 7 Oct 2026

Does Writing with Language Models Reduce Content Diversity?

arXiv 2309.05196 ↗First posted 11 Sep 2023cs.CL cs.CY cs.HCcited by 190

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) have led to a surge in collaborative writing with model assistance. As different users incorporate suggestions from the same model, there is a risk of decreased diversity in the produced content, potentially limiting diverse perspectives in public discourse. In this work, we measure the impact of co-writing on diversity via a controlled experiment, where users write argumentative essays in three setups -- using a base LLM (GPT3), a feedback-tuned LLM (InstructGPT), and writing without model help. We develop a set of diversity metrics and find that writing with InstructGPT (but not the GPT3) results in a statistically significant reduction in diversity. Specifically, it increases the similarity between the writings of different authors and reduces the overall lexical and content diversity. We additionally find that this effect is mainly attributable to InstructGPT contributing less diverse text to co-written essays. In contrast, the user-contributed text remains unaffected by model collaboration. This suggests that the recent improvement in generation quality from adapting models to human feedback might come at the cost of more homogeneous and less diverse content.

Key facts. Cited by 190 papers in the map (data through 7 Oct 2026). Builds on 33 of its 67 references that are in the map. First posted on arXiv 11 Sep 2023.

Builds on

33 of 67 references.

References · 33

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,462
  3. 2021On the Opportunities and Risks of Foundation Modelscited by 3,961
  4. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  5. 2020BLEURT: Learning Robust Metrics for Text Generationcited by 1,196
  6. 2023Whose Opinions Do Language Models Reflect?cited by 759
  7. 2023Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedbackcited by 743
  8. 2022Contrastive Decoding: Open-ended Text Generation as Optimizationcited by 598
  9. 2018Texygen: A Benchmarking Platform for Text Generation Modelscited by 484
  10. 2021MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontierscited by 466
  11. 2023Towards Measuring the Representation of Subjective Global Opinions in Language Modelscited by 424
  12. 2022News Summarization and Evaluation in the Era of GPT-3cited by 324
21 more
  1. 2022CoAuthor: Designing a Human-AI Collaborative Writing Dataset for Exploring Language Model Capabilitiescited by 245
  2. 2022Co-Writing Screenplays and Theatre Scripts with Language Models: An Evaluation by Industry Professionalscited by 191
  3. 2022Repairing the Cracked Foundation: A Survey of Obstacles in Evaluation Practices for Generated Textcited by 154
  4. 2023Co-Writing with Opinionated Language Models Affects Users' Viewscited by 145
  5. 2020Evaluating the Evaluation of Diversity in Natural Language Generationcited by 127
  6. 2022Locally Typical Samplingcited by 108
  7. 2021Algorithmic Monoculture and Social Welfarecited by 98
  8. 2023Evaluating the Social Impact of Generative AI Systems in Systems and Societycited by 91
  9. 2019Do Massively Pretrained Language Models Make Better Storytellers?cited by 90
  10. 2022Evaluating Human-Language Model Interactioncited by 90
  11. 2022Picking on the Same Person: Does Algorithmic Monoculture lead to Outcome Homogenization?cited by 89
  12. 2022Help me write a poem: Instruction Tuning as a Vehicle for Collaborative Poetry Writingcited by 71
  13. 2022Creative Writing with an AI-Powered Writing Assistant: Perspectives from Professional Writerscited by 69
  14. 2021The Impact of Multiple Parallel Phrase Suggestions on Email Input and Composition Behaviour of Native and Non-Native English Writerscited by 61
  15. 2022Data Feedback Loops: Model-driven Amplification of Dataset Biasescited by 59
  16. 2020STORIUM: A Dataset and Evaluation Platform for Machine-in-the-Loop Story Generationcited by 56
  17. 2023Rethinking Model Evaluation as Narrowing the Socio-Technical Gapcited by 56
  18. 2022Interacting with next-phrase suggestions: How suggestion systems aid and influence the cognitive processes of writingcited by 27
  19. 2021Divergence Frontiers for Generative Models: Sample Complexity, Quantization Effects, and Frontier Integralscited by 18
  20. 2020Langsmith: An Interactive Academic Text Revision Systemcited by 13
  21. 2021Machine-in-the-Loop Rewriting for Creative Image Captioningcited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 7; 2024: 43; 2025: 55; 2026: 85.

190 papers.

Sort:

By area:

  1. 2024Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researcherscited by 348
  2. 2024Homogenization Effects of Large Language Models on Human Creative Ideationcited by 127
  3. 2024Survey of Cultural Awareness in Language Models: Text and Beyondcited by 101
  4. 2023The Curious Decline of Linguistic Diversity: Training Language Models on Synthetic Textcited by 97
  5. 2025Jointly Reinforcing Diversity and Quality in Language Model Generationscited by 79
  6. 2024A Design Space for Intelligent and Interactive Writing Assistantscited by 76
  7. 2024Delving into LLM-assisted writing in biomedical publications through excess vocabularycited by 75
  8. 2024LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Playcited by 65
  9. 2023Composite Backdoor Attacks Against Large Language Modelscited by 63
  10. 2024On the Diversity of Synthetic Data and its Impact on Training Large Language Modelscited by 56
  11. 2023Bridging the Novice-Expert Gap via Models of Decision-Making: A Case Study on Remediating Math Mistakescited by 51
  12. 2025Evaluating the Diversity and Quality of LLM Generated Contentcited by 51
38 more
  1. 2024Standardizing the Measurement of Text Diversity: A Tool and a Comparative Analysis of Scorescited by 50
  2. 2024Can AI writing be salvaged? Mitigating Idiosyncrasies and Improving Human-AI Alignment in the Writing Process through Editscited by 49
  3. 2024Collapse or Thrive? Perils and Promises of Synthetic Data in a Self-Generating Worldcited by 49
  4. 2025Diverse Preference Optimizationcited by 48
  5. 2025The Best Instruction-Tuning Data are Those That Fitcited by 48
  6. 2025Base Models Beat Aligned Models at Randomness and Creativitycited by 48
  7. 2024Echoes in AI: Quantifying lack of plot diversity in LLM outputscited by 47
  8. 2024Detection and Measurement of Syntactic Templates in Generated Textcited by 44
  9. 2024Beyond Model Collapse: Scaling Up with Synthesized Data Requires Verificationcited by 43
  10. 2024Human Creativity in the Age of LLMs: Randomized Experiments on Divergent and Convergent Thinkingcited by 43
  11. 2024Human-LLM Compound System for Scientific Ideation through Facet Recombination and Novelty Evaluationcited by 39
  12. 2025Modifying Large Language Model Post-Training for Diverse Creative Writingcited by 39
  13. 2024Benchmarking Linguistic Diversity of Large Language Modelscited by 37
  14. 2023Beyond the Chat: Executable and Verifiable Text-Editing with LLMscited by 35
  15. 2025Position: Model Collapse Does Not Mean What You Thinkcited by 32
  16. 2024Divergent Creativity in Humans and Large Language Modelscited by 29
  17. 2024A Robot Walks into a Bar: Can Language Models Serve as Creativity Support Tools for Comedy? An Evaluation of LLMs' Humour Alignment with Comedianscited by 29
  18. 2024Debate-to-Write: A Persona-Driven Multi-Agent Framework for Diverse Argument Generationcited by 29
  19. 2025LLM Probability Concentration: How Alignment Shrinks the Generative Horizoncited by 26
  20. 2024Towards a Scalable Reference-Free Evaluation of Generative Modelscited by 25
  21. 2025Multilingual Prompting for Improving LLM Generation Diversitycited by 25
  22. 2025Diverse Preference Learning for Capabilities and Alignmentcited by 25
  23. 2025AI-Slop to AI-Polish? Aligning Language Models through Edit-Based Writing Rewards and Test-time Computationcited by 24
  24. 2025Creative Preference Optimizationcited by 24
  25. 2023How Far Can We Extract Diverse Perspectives from Large Language Models?cited by 22
  26. 2024How AI Ideas Affect the Creativity, Diversity, and Evolution of Human Ideas: Evidence From a Large, Dynamic Experimentcited by 22
  27. 2024Generative Monoculture in Large Language Modelscited by 21
  28. 2024Evaluating Creative Short Story Generation in Humans and Large Language Modelscited by 21
  29. 2024Reading Subtext: Evaluating Large Language Models on Short Story Summarization with Writerscited by 19
  30. 2025Post-training Large Language Models for Diverse High-Quality Responsescited by 19
  31. 2026Brittlebench: Quantifying LLM robustness via prompt sensitivitycited by 19
  32. 2024GhostWriter: Augmenting Collaborative Human-AI Writing Experiences Through Personalization and Agencycited by 18
  33. 2024Plurals: A System for Guiding LLMs Via Simulated Social Ensemblescited by 17
  34. 2024Regurgitative Training: The Value of Real Data in Training Large Language Modelscited by 15
  35. 2024Measuring Non-Adversarial Reproduction of Training Data in Large Language Modelscited by 14
  36. 2025What and Whose Knowledge? Measuring Epistemic Diversity in Large Language Modelscited by 14
  37. 2024Synergy-of-Thoughts: Eliciting Efficient Reasoning in Hybrid Language Modelscited by 13
  38. 2024Collage is the New Writing: Exploring the Fragmentation of Text and User Interfaces in AI Toolscited by 13

50 most cited of 190 All 190 (JSON)