mlsanityData to 7 Oct 2026

Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection

arXiv 2004.07667 ↗First posted 16 Apr 2020cs.CL cs.LGcited by 442

Not on the map. Why?

Cited by 22 in LLM safety and security, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

The ability to control for the kinds of information encoded in neural representation has a variety of use cases, especially in light of the challenge of interpreting these models. We present Iterative Null-space Projection (INLP), a novel method for removing information from neural representations. Our method is based on repeated training of linear classifiers that predict a certain property we aim to remove, followed by projection of the representations on their null-space. By doing so, the classifiers become oblivious to that target property, making it hard to linearly separate the data according to it. While applicable for multiple uses, we evaluate our method on bias and fairness use-cases, and show that our method is able to mitigate bias in word embeddings, as well as to increase fairness in a setting of multi-class classification.

Key facts. Cited by 442 papers in the map (data through 7 Oct 2026). First posted on arXiv 16 Apr 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 28; 2022: 60; 2023: 73; 2024: 82; 2025: 73; 2026: 126.

442 papers.

Sort:

By area:

  1. 2024Refusal in Language Models Is Mediated by a Single Directioncited by 960
  2. 2023Bias and Fairness in Large Language Models: A Surveycited by 626
  3. 2024Mechanistic Interpretability for AI Safety -- A Reviewcited by 415
  4. 2024Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Modelscited by 412
  5. 2021Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLPcited by 336
  6. 2021Towards Understanding and Mitigating Social Biases in Language Modelscited by 294
  7. 2023LEACE: Perfect linear concept erasure in closed formcited by 254
  8. 2023Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representationscited by 197
  9. 2023Causal Abstraction: A Theoretical Foundation for Mechanistic Interpretabilitycited by 194
  10. 2021Causal Inference in Natural Language Processing: Estimation, Prediction, Interpretation and Beyondcited by 177
  11. 2023Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attackscited by 156
  12. 2024Improving Instruction-Following in Language Models through Activation Steeringcited by 149
38 more
  1. 2023Inspecting and Editing Knowledge Representations in Language Modelscited by 140
  2. 2022Towards Faithful Model Explanation in NLP: A Surveycited by 129
  3. 2021Harms of Gender Exclusivity and Challenges in Non-Binary Representation in Language Technologiescited by 128
  4. 2021Debiasing Pre-trained Contextualised Embeddingscited by 107
  5. 2021Inducing Causal Structure for Interpretable Neural Networkscited by 107
  6. 2023Interpretability at Scale: Identifying Causal Mechanisms in Alpacacited by 102
  7. 2022Linear Adversarial Concept Erasurecited by 98
  8. 2022What Artificial Neural Networks Can Tell Us About Human Language Acquisitioncited by 97
  9. 2021An Empirical Survey of the Effectiveness of Debiasing Techniques for Pre-trained Language Modelscited by 96
  10. 2023A Survey on Fairness in Large Language Modelscited by 91
  11. 2023Do Models Explain Themselves? Counterfactual Simulatability of Natural Language Explanationscited by 85
  12. 2024A Survey on Multilingual Large Language Models: Corpora, Alignment, and Biascited by 82
  13. 2021On Measures of Biases and Harms in NLPcited by 81
  14. 2024RAVEL: Evaluating Interpretability Methods on Disentangling Language Model Representationscited by 77
  15. 2024Bias in Large Language Models: Origin, Evaluation, and Mitigationcited by 77
  16. 2021Contrastive Explanations for Model Interpretabilitycited by 75
  17. 2023Word Embeddings Are Steers for Language Modelscited by 70
  18. 2023Fast Model Debias with Machine Unlearningcited by 70
  19. 2023SeeGULL: A Stereotype Benchmark with Broad Geo-Cultural Coverage Leveraging Generative Modelscited by 66
  20. 2025Convergent Linear Representations of Emergent Misalignmentcited by 66
  21. 2022Theories of "Gender" in NLP Bias Researchcited by 62
  22. 2024Robust LLM safeguarding via refusal feature adversarial trainingcited by 60
  23. 2024Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineeringcited by 60
  24. 2021Counterfactual Interventions Reveal the Causal Effect of Relative Clause Representations on Agreement Predictioncited by 58
  25. 2022Probing for the Usage of Grammatical Numbercited by 57
  26. 2024Fairness in Large Language Models: A Taxonomic Surveycited by 57
  27. 2022Entropy-based Attention Regularization Frees Unintended Bias Mitigation from Listscited by 54
  28. 2022CEBaB: Estimating the Causal Effects of Real-World Concepts on NLP Model Behaviorcited by 54
  29. 2023The Curious Case of Hallucinatory (Un)answerability: Finding Truths in the Hidden States of Over-Confident Large Language Modelscited by 54
  30. 2024Spectral Editing of Activations for Large Language Model Alignmentcited by 54
  31. 2022Fair and Optimal Classification via Post-Processingcited by 50
  32. 2021Diverse Adversaries for Mitigating Bias in Trainingcited by 48
  33. 2022Debiasing isn't enough! -- On the Effectiveness of Debiasing MLMs and their Social Biases in Downstream Taskscited by 48
  34. 2025Do Language Models Use Their Depth Efficiently?cited by 48
  35. 2022Kernelized Concept Erasurecited by 47
  36. 2024Representation Surgery: Theory and Practice of Affine Steeringcited by 47
  37. 2024Recurrent Neural Networks Learn to Store and Generate Sequences using Non-Linear Representationscited by 45
  38. 2023Break It Down: Evidence for Structural Compositionality in Neural Networkscited by 44

50 most cited of 442 All 442 (JSON)