mlsanityData to 7 Oct 2026

Debiasing Algorithm through Model Adaptation

arXiv 2310.18913 ↗First posted 29 Oct 2023cs.CL cs.AI stat.MLcited by 24

Not on the map. Why?

Abstract (arXiv)

Large language models are becoming the go-to solution for the ever-growing number of tasks. However, with growing capacity, models are prone to rely on spurious correlations stemming from biases and stereotypes present in the training data. This work proposes a novel method for detecting and mitigating gender bias in language models. We perform causal analysis to identify problematic model components and discover that mid-upper feed-forward layers are most prone to convey bias. Based on the analysis results, we intervene in the model by applying a linear projection to the weight matrices of these layers. Our titular method, DAMA, significantly decreases bias as measured by diverse metrics while maintaining the model's performance on downstream tasks. We release code for our method and models, which retrain LLaMA's state-of-the-art performance while being significantly less biased.

Key facts. Cited by 24 papers in the map (data through 7 Oct 2026). Builds on 31 of its 40 references that are in the map. First posted on arXiv 29 Oct 2023.

Builds on

31 of 40 references.

References · 31

  1. 2017Attention Is All You Needcited by 59,932
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  3. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  4. 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,710
  5. 2016Pointer Sentinel Mixture Modelscited by 2,992
  6. 2022Locating and Editing Factual Associations in GPTcited by 2,609
  7. 2018Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answeringcited by 2,114
  8. 2020Transformer Feed-Forward Layers Are Key-Value Memoriescited by 1,307
  9. 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,304
  10. 2022Mass-Editing Memory in a Transformercited by 976
  11. 2020Language (Technology) is Power: A Critical Survey of "Bias" in NLPcited by 951
  12. 2020StereoSet: Measuring stereotypical bias in pretrained language modelscited by 911
19 more
  1. 2020CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Modelscited by 703
  2. 2018Gender Bias in Coreference Resolution: Evaluation and Debiasing Methodscited by 657
  3. 2021Editing Factual Knowledge in Language Modelscited by 548
  4. 2021Fast Model Editing at Scalecited by 491
  5. 2018Gender Bias in Coreference Resolutioncited by 438
  6. 2013Direct and Indirect Effectscited by 276
  7. 2023LEACE: Perfect linear concept erasure in closed formcited by 254
  8. 2019Evaluating Gender Bias in Machine Translationcited by 247
  9. 2018Gender Bias in Neural Natural Language Processingcited by 210
  10. 2019Counterfactual Data Augmentation for Mitigating Gender Stereotypes in Languages with Rich Morphologycited by 200
  11. 2020Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Biascited by 183
  12. 2021Gender Bias in Machine Translationcited by 116
  13. 2021Sustainable Modular Debiasing of Language Modelscited by 98
  14. 2022Linear Adversarial Concept Erasurecited by 98
  15. 2021A Survey on Gender Bias in Natural Language Processingcited by 89
  16. 2023A Trip Towards Fairness: Bias and De-Biasing in Large Language Modelscited by 32
  17. 2022AdapterBias: Parameter-efficient Token-dependent Representation Shift for Adapters in NLP Taskscited by 29
  18. 2022Don't Forget About Pronouns: Removing Gender Bias in Language Models Without Losing Factual Gender Informationcited by 16
  19. 2023An Empirical Analysis of Parameter-Efficient Methods for Debiasing Pre-Trained Language Modelscited by 16

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 13; 2025: 7; 2026: 4.

24 papers.

Sort:

By area:

  1. 2024Safeguarding Large Language Models: A Surveycited by 73
  2. 2024Navigating the Risks: A Survey of Security, Privacy, and Ethics Threats in LLM-Based Agentscited by 63
  3. 2024Building Guardrails for Large Language Modelscited by 61
  4. 2024Fairness in Large Language Models: A Taxonomic Surveycited by 57
  5. 2024Attribution Analysis Meets Model Editing: Advancing Knowledge Correction in Vision Language Models with VisEditcited by 14
  6. 2024Lifelong Knowledge Editing for Vision Language Models with Low-Rank Mixture-of-Expertscited by 14
  7. 2024Look Within, Why LLMs Hallucinate: A Causal Perspectivecited by 11
  8. 2024Potential and Challenges of Model Editing for Social Debiasingcited by 10
  9. 2024Mitigating Biases of Large Language Models in Stance Detection with Counterfactual Augmented Calibrationcited by 5
  10. 2024Is Free Self-Alignment Possible?cited by 4
  11. 2024Mitigating Gender Bias in Code Large Language Models via Model Editingcited by 3
  12. 2025Preserving Task-Relevant Information Under Linear Concept Removalcited by 3
12 more
  1. 2025KLAAD: Refining Attention Mechanisms to Reduce Societal Bias in Generative Language Modelscited by 3
  2. 2024Runtime Monitoring and Enforcement of Conditional Fairness in Generative AIscited by 2
  3. 2025Investigating Hallucination in Conversations for Low Resource Languagescited by 2
  4. 2025BiasFreeBench: a Benchmark for Mitigating Bias in Large Language Model Responsescited by 2
  5. 2026Attention, May I Have Your Decision? Localizing Generative Choices in Diffusion Modelscited by 2
  6. 2026Neuron-Level Interventions for Gendered and Gender-Neutral Generation in Language Modelscited by 2
  7. 2024Gender Bias of LLM in Economics: An Existentialism Perspectivecited by 1
  8. 2025Advertising in AI systems: Society must be vigilantcited by 1
  9. 2025A Critical Review of Predominant Bias in Neural Networkscited by 0
  10. 2025Debiasing Large Language Models in Thai Political Stance Detection via Counterfactual Calibrationcited by 0
  11. 2026DeFrame: Debiasing Large Language Models Against Framing Effectscited by 0
  12. 2026GKnow: Measuring the Entanglement of Gender Bias and Factual Gendercited by 0