# Shielded Representations: Protecting Sensitive Attributes Through Iterative Gradient-Based Projection

arXiv: https://arxiv.org/abs/2305.10204 · HTML: https://mlsanity.com/paper/2305.10204/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 41 references that are in the map.
- First posted on arXiv 17 May 2023.

## Builds on (most cited first, 27 of 27)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Semantics derived automatically from language corpora contain human-like biases](https://mlsanity.com/paper/1608.07187/) (2016)
- [Probing Classifiers: Promises, Shortcomings, and Advances](https://mlsanity.com/paper/2102.12452/) (2021)
- [Mitigating Unwanted Biases with Adversarial Learning](https://mlsanity.com/paper/1801.07593/) (2018)
- [Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods](https://mlsanity.com/paper/1804.06876/) (2018)
- [Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection](https://mlsanity.com/paper/2004.07667/) (2020)
- [Gender Bias in Coreference Resolution](https://mlsanity.com/paper/1804.09301/) (2018)
- [Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Setting](https://mlsanity.com/paper/1901.09451/) (2019)
- [Lipstick on a Pig: Debiasing Methods Cover up Systematic Gender Biases in Word Embeddings But do not Remove Them](https://mlsanity.com/paper/1903.03862/) (2019)
- [Information-Theoretic Probing with Minimum Description Length](https://mlsanity.com/paper/2003.12298/) (2020)
- [Demographic Dialectal Variation in Social Media: A Case Study of African-American English](https://mlsanity.com/paper/1608.08868/) (2016)
- [SimLex-999: Evaluating Semantic Models with (Genuine) Similarity Estimation](https://mlsanity.com/paper/1408.3456/) (2014)
- [Adversarial Removal of Demographic Attributes from Text Data](https://mlsanity.com/paper/1808.06640/) (2018)
- [Data Decisions and Theoretical Implications when Adversarially Learning Fair Representations](https://mlsanity.com/paper/1707.00075/) (2017)
- [Controllable Invariance through Adversarial Feature Learning](https://mlsanity.com/paper/1705.11122/) (2017)
- [Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm](https://mlsanity.com/paper/1708.00524/) (2017)
- [Linear Adversarial Concept Erasure](https://mlsanity.com/paper/2201.12091/) (2022)
- [What's in a Name? Reducing Bias in Bios without Access to Protected Attributes](https://mlsanity.com/paper/1904.05233/) (2019)
- [Kernelized Concept Erasure](https://mlsanity.com/paper/2201.12191/) (2022)
- [Probing Classifiers are Unreliable for Concept Removal and Detection](https://mlsanity.com/paper/2207.04153/) (2022)
- [Double-Hard Debias: Tailoring Word Embeddings for Gender Bias Mitigation](https://mlsanity.com/paper/2005.00965/) (2020)
- [Gold Doesn't Always Glitter: Spectral Removal of Linear and Nonlinear Guarded Attribute Information](https://mlsanity.com/paper/2203.07893/) (2022)
- [Choose Your Lenses: Flaws in Gender Bias Evaluation](https://mlsanity.com/paper/2210.11471/) (2022)
- [Debiasing Methods in Natural Language Understanding Make Bias More Accessible](https://mlsanity.com/paper/2109.04095/) (2021)
- [How Gender Debiasing Affects Internal Model Representations, and Why It Matters](https://mlsanity.com/paper/2204.06827/) (2022)

## Built on by (most cited first, 25 of 25)

- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models](https://mlsanity.com/paper/2403.19647/) (2024)
- [A Survey on Fairness in Large Language Models](https://mlsanity.com/paper/2308.10149/) (2023)
- [Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering](https://mlsanity.com/paper/2410.15999/) (2024)
- [Spectral Editing of Activations for Large Language Model Alignment](https://mlsanity.com/paper/2405.09719/) (2024)
- [Steering Out-of-Distribution Generalization with Concept Ablation Fine-Tuning](https://mlsanity.com/paper/2507.16795/) (2025)
- [GenderBias-VL: Benchmarking Gender Bias in Vision Language Models via Counterfactual Probing](https://mlsanity.com/paper/2407.00600/) (2024)
- [The Quest for the Right Mediator: Surveying Mechanistic Interpretability Through the Lens of Causal Mediation Analysis](https://mlsanity.com/paper/2408.01416/) (2024)
- [Deconstructing The Ethics of Large Language Models from Long-standing Issues to New-emerging Dilemmas: A Survey](https://mlsanity.com/paper/2406.05392/) (2024)
- [BiasEdit: Debiasing Stereotyped Language Models via Model Editing](https://mlsanity.com/paper/2503.08588/) (2025)
- [On the Scaling Laws of Geographical Representation in Language Models](https://mlsanity.com/paper/2402.19406/) (2024)
- [AGR: Age Group fairness Reward for Bias Mitigation in LLMs](https://mlsanity.com/paper/2409.04340/) (2024)
- [Tackling Bias in Pre-trained Language Models: Current Trends and Under-represented Societies](https://mlsanity.com/paper/2312.01509/) (2023)
- [Precise In-Parameter Concept Erasure in Large Language Models](https://mlsanity.com/paper/2505.22586/) (2025)
- [Spectral Attention Steering for Prompt Highlighting](https://mlsanity.com/paper/2603.01281/) (2026)
- [Focus On This, Not That! Steering LLMs with Adaptive Feature Specification](https://mlsanity.com/paper/2410.22944/) (2024)
- [Leveraging Prototypical Representations for Mitigating Social Bias without Demographic Information](https://mlsanity.com/paper/2403.09516/) (2024)
- [The Life Cycle of Large Language Models: A Review of Biases in Education](https://mlsanity.com/paper/2407.11203/) (2024)
- [MultiContrievers: Analysis of Dense Retrieval Representations](https://mlsanity.com/paper/2402.15925/) (2024)
- [FALCON: Fine-grained Activation Manipulation by Contrastive Orthogonal Unalignment for Large Language Model](https://mlsanity.com/paper/2502.01472/) (2025)
- [Multimodal Large Language Models for Enhanced Traffic Safety: A Comprehensive Review and Future Trends](https://mlsanity.com/paper/2504.16134/) (2025)
- [Understanding Empirical Unlearning with Combinatorial Interpretability](https://mlsanity.com/paper/2602.19215/) (2026)
- [MANCE: Manifold Aware Concept Erasure](https://mlsanity.com/paper/2607.03973/) (2026)
- [Fair on the Surface? Benchmarking Hidden-Output Fairness Gaps in LLM Recommenders](https://mlsanity.com/paper/2608.08284/) (2026)
- [Measurement-Gated Provenance Attenuation for Frozen EEG Representations](https://mlsanity.com/paper/2609.32889/) (2026)

## Abstract (arXiv)

Natural language processing models tend to learn and encode social biases present in the data. One popular approach for addressing such biases is to eliminate encoded information from the model's representations. However, current methods are restricted to removing only linearly encoded information. In this work, we propose Iterative Gradient-Based Projection (IGBP), a novel method for removing non-linear encoded concepts from neural representations. Our method consists of iteratively training neural classifiers to predict a particular attribute we seek to eliminate, followed by a projection of the representation on a hypersurface, such that the classifiers become oblivious to the target attribute. We evaluate the effectiveness of our method on the task of removing gender and race information as sensitive attributes. Our results demonstrate that IGBP is effective in mitigating bias through intrinsic and extrinsic evaluations, with minimal impact on downstream task accuracy.