mlsanityData to 7 Oct 2026

Identity-Robust Language Model Generation via Content Integrity Preservation

arXiv 2601.09141 ↗First posted 14 Jan 2026cs.CLcited by 1

Not on the map. Why?

Abstract (arXiv)

Large Language Model (LLM) outputs often vary across user sociodemographic attributes, leading to disparities in factual accuracy, utility, and safety, even for objective questions where demographic information is irrelevant. Unlike prior work on stereotypical or representational bias, this paper studies identity-dependent degradation of core response quality. We show empirically that such degradation arises from biased generation behavior, despite factual knowledge being robustly encoded across identities. Motivated by this mismatch, we propose a lightweight, training-free framework for identity-robust generation that selectively neutralizes non-critical identity information while preserving semantically essential attributes, thus maintaining output content integrity. Experiments across four benchmarks and 18 sociodemographic identities demonstrate an average 77% reduction in identity-dependent bias compared to vanilla prompting and a 45% reduction relative to prompt-based defenses. Our work addresses a critical gap in mitigating the impact of user identity cues in prompts on core generation quality.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 29 of its 47 references that are in the map. First posted on arXiv 14 Jan 2026.

Builds on

29 of 47 references.

References · 29

  1. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,919
  2. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,758
  3. 2023Inference-Time Intervention: Eliciting Truthful Answers from a Language Modelcited by 1,259
  4. 2023The Internal State of an LLM Knows When It's Lyingcited by 644
  5. 2024WildChat: 1M ChatGPT Interaction Logs in the Wildcited by 598
  6. 2020AmbigQA: Answering Ambiguous Open-domain Questionscited by 385
  7. 2024LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinationscited by 230
  8. 2023Sensitivity, Performance, Robustness: Deconstructing the Effect of Sociodemographic Promptingcited by 80
  9. 2024Self-Debiasing Large Language Models: Zero-Shot Recognition and Reduction of Stereotypescited by 58
  10. 2024Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrievalcited by 49
  11. 2024Thinking Fair and Slow: On the Efficacy of Structured Prompts for Debiasing Language Modelscited by 44
  12. 2024Decision-Making Behavior Evaluation Framework for LLMs under Uncertain Contextcited by 41
17 more
  1. 2024Who's asking? User personas and the mechanics of latent misalignmentcited by 40
  2. 2024UniBias: Unveiling and Mitigating LLM Bias through Internal Attention and FFN Manipulationcited by 30
  3. 2024LLMs are Biased Teachers: Evaluating LLM Bias in Personalized Educationcited by 28
  4. 2024Exploring Safety-Utility Trade-Offs in Personalized Language Modelscited by 26
  5. 2025Reading Between the Prompts: How Stereotypes Shape LLM's Implicit Personalizationcited by 24
  6. 2024Reinforcement Learning from Multi-role Debates as Feedback for Bias Mitigation in LLMscited by 22
  7. 2025Aligned but Blind: Alignment Increases Implicit Bias by Reducing Awareness of Racecited by 15
  8. 2023Filter bubbles and affective polarization in user-personalized large language model outputscited by 6
  9. 2024The Male CEO and the Female Assistant: Evaluation and Mitigation of Gender Biases in Text-To-Image Generation of Dual Subjectscited by 6
  10. 2025Mitigating Biases in Language Models via Bias Unlearningcited by 6
  11. 2024The Generation Gap: Exploring Age Bias in the Value Systems of Large Language Modelscited by 5
  12. 2025Where Fact Ends and Fairness Begins: Redefining AI Bias Evaluation through Cognitive Biasescited by 5
  13. 2025Assumed Identities: Quantifying Gender Bias in Machine Translation of Gender-Ambiguous Occupational Termscited by 5
  14. 2025On the Mutual Influence of Gender and Occupation in LLM Representationscited by 4
  15. 2025Mitigating Gender Bias via Fostering Exploratory Thinking in LLMscited by 4
  16. 2025LLM Bias Detection and Mitigation through the Lens of Desired Distributionscited by 4
  17. 2025Bias Analysis and Mitigation through Protected Attribute Detection and Regard Classificationcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Mitigating LLM biases toward spurious social contexts using direct preference optimizationcited by 0