# Debiasing Without Protected Attributes: Latent Concept Erasure from Textual Profiles

arXiv: https://arxiv.org/abs/2606.12088 · HTML: https://mlsanity.com/paper/2606.12088/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 30 references that are in the map.
- First posted on arXiv 10 Jun 2026.

## Builds on (most cited first, 20 of 20)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection](https://mlsanity.com/paper/2004.07667/) (2020)
- [Bias in Bios: A Case Study of Semantic Representation Bias in a High-Stakes Setting](https://mlsanity.com/paper/1901.09451/) (2019)
- [LEACE: Perfect linear concept erasure in closed form](https://mlsanity.com/paper/2306.03819/) (2023)
- [All-but-the-Top: Simple and Effective Postprocessing for Word Representations](https://mlsanity.com/paper/1702.01417/) (2017)
- [Adversarial Removal of Demographic Attributes from Text Data](https://mlsanity.com/paper/1808.06640/) (2018)
- [Learning to Ask Good Questions: Ranking Clarification Questions using Neural Expected Value of Perfect Information](https://mlsanity.com/paper/1805.04655/) (2018)
- [Isambard-AI: a leadership class supercomputer optimised specifically for Artificial Intelligence](https://mlsanity.com/paper/2410.11199/) (2024)
- [Privacy-preserving Neural Representations of Text](https://mlsanity.com/paper/1808.09408/) (2018)
- [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)
- [Text-to-SQL in the Wild: A Naturally-Occurring Dataset Based on Stack Exchange Data](https://mlsanity.com/paper/2106.05006/) (2021)
- [Gold Doesn't Always Glitter: Spectral Removal of Linear and Nonlinear Guarded Attribute Information](https://mlsanity.com/paper/2203.07893/) (2022)
- [Proxy Fairness](https://mlsanity.com/paper/1806.11212/) (2018)
- [Reading Between the Prompts: How Stereotypes Shape LLM's Implicit Personalization](https://mlsanity.com/paper/2505.16467/) (2025)
- [BLIND: Bias Removal With No Demographics](https://mlsanity.com/paper/2212.10563/) (2022)
- [Sparse Activation Editing for Reliable Instruction Following in Narratives](https://mlsanity.com/paper/2505.16505/) (2025)
- [Differentiable Faithfulness Alignment for Cross-Model Circuit Transfer](https://mlsanity.com/paper/2604.24302/) (2026)
- [Iterative Multilingual Spectral Attribute Erasure](https://mlsanity.com/paper/2506.11244/) (2025)

## Abstract (arXiv)

Most fairness research in NLP assumes direct access to protected attributes such as gender, race, or nationality. In practice, however, such information is often unavailable due to privacy constraints, missing metadata, or legal restrictions, even though models may infer it from indirect textual cues. This raises a key question: can debiasing succeed without direct access to sensitive attributes? We propose H-SAL, which performs post-hoc concept and attribute erasure using self-description text as an implicit debiasing signal. To support this setting, we introduce a multi-domain Stack Exchange-based fairness benchmark for helpfulness prediction that includes both explicit and implicit signals, enabling comparison between standard debiasing with protected labels and debiasing without access to sensitive information. Across encoder and decoder-only language models, we find that implicit self-description often matches or outperforms explicit-label-based debiasing. Our results broaden representation-level fairness research and provide a new benchmark for studying debiasing under realistic data constraints.