# FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering

arXiv: https://arxiv.org/abs/2504.14492 · HTML: https://mlsanity.com/paper/2504.14492/

## Key facts

- Cited by 40 papers in the map (data through 14 Oct 2026).
- Placed in Probing deception, bias and beliefs, one of 13 directions in Mechanistic interpretability.
- Builds on 62 of its 78 references that are in the map.
- First posted on arXiv 20 Apr 2025.

## Placed in

- Mechanistic interpretability / [Probing deception, bias and beliefs](https://mlsanity.com/mechinterp/probing-deception-bias-and-beliefs/)

## Builds on (most cited first, 30 of 62)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering](https://mlsanity.com/paper/1809.02789/) (2018)
- [Representation Engineering: A Top-Down Approach to AI Transparency](https://mlsanity.com/paper/2310.01405/) (2023)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Refusal in Language Models Is Mediated by a Single Direction](https://mlsanity.com/paper/2406.11717/) (2024)
- [Toy Models of Superposition](https://mlsanity.com/paper/2209.10652/) (2022)
- [CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models](https://mlsanity.com/paper/2010.00133/) (2020)
- [BBQ: A Hand-Built Bias Benchmark for Question Answering](https://mlsanity.com/paper/2110.08193/) (2021)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [The Woman Worked as a Babysitter: On Biases in Language Generation](https://mlsanity.com/paper/1909.01326/) (2019)
- [Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection](https://mlsanity.com/paper/2004.07667/) (2020)
- [DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts](https://mlsanity.com/paper/2105.03023/) (2021)
- [BOLD: Dataset and Metrics for Measuring Biases in Open-Ended Language Generation](https://mlsanity.com/paper/2101.11718/) (2021)
- [Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLP](https://mlsanity.com/paper/2103.00453/) (2021)
- [Prompting GPT-3 To Be Reliable](https://mlsanity.com/paper/2210.09150/) (2022)
- [Measuring and Reducing Gendered Correlations in Pre-trained Models](https://mlsanity.com/paper/2010.06032/) (2020)
- [Gender Bias in Neural Natural Language Processing](https://mlsanity.com/paper/1807.11714/) (2018)
- [Gender Bias in Contextualized Word Embeddings](https://mlsanity.com/paper/1904.03310/) (2019)
- [Societal Biases in Language Generation: Progress and Challenges](https://mlsanity.com/paper/2105.04054/) (2021)

## Built on by (most cited first, 30 of 40)

- [SAEs Are Good for Steering -- If You Select the Right Features](https://mlsanity.com/paper/2505.20063/) (2025)
- [Angular Steering: Behavior Control via Rotation in Activation Space](https://mlsanity.com/paper/2510.26243/) (2025)
- [SAE-SSV: Supervised Steering in Sparse Representation Spaces for Reliable Control of Language Models](https://mlsanity.com/paper/2505.16188/) (2025)
- [Silenced Biases: The Dark Side LLMs Learned to Refuse](https://mlsanity.com/paper/2511.03369/) (2025)
- [Guiding Giants: Lightweight Controllers for Weighted Activation Steering in LLMs](https://mlsanity.com/paper/2505.20309/) (2025)
- [ARGUS: Defending Against Multimodal Indirect Prompt Injection via Steering Instruction-Following Behavior](https://mlsanity.com/paper/2512.05745/) (2025)
- [BiasFilter: An Inference-Time Debiasing Framework for Large Language Models](https://mlsanity.com/paper/2505.23829/) (2025)
- [Selective Steering: Norm-Preserving Control Through Discriminative Layer Selection](https://mlsanity.com/paper/2601.19375/) (2026)
- [C2PO: Diagnosing and Disentangling Bias Shortcuts in LLMs](https://mlsanity.com/paper/2512.23430/) (2025)
- [The Hidden Signal of Verifier Strictness: Controlling and Improving Step-Wise Verification via Selective Latent Steering](https://mlsanity.com/paper/2605.20745/) (2026)
- [BiasGuard: A Reasoning-enhanced Bias Detection Tool For Large Language Models](https://mlsanity.com/paper/2504.21299/) (2025)
- [DSO: Direct Steering Optimization for Bias Mitigation](https://mlsanity.com/paper/2512.15926/) (2025)
- [Self-Debias: Self-correcting for Debiasing Large Language Models](https://mlsanity.com/paper/2604.08243/) (2026)
- [Where Steering Signals Come From: Activation Source Selection in Activation Steering](https://mlsanity.com/paper/2607.25270/) (2026)
- [BiasFreeBench: a Benchmark for Mitigating Bias in Large Language Model Responses](https://mlsanity.com/paper/2510.00232/) (2025)
- [Robust Preference Alignment via Directional Neighborhood Consensus](https://mlsanity.com/paper/2510.20498/) (2025)
- [Can SAEs reveal and mitigate racial biases of LLMs in healthcare?](https://mlsanity.com/paper/2511.00177/) (2025)
- [Knowing without Acting: The Disentangled Geometry of Safety Mechanisms in Large Language Models](https://mlsanity.com/paper/2603.05773/) (2026)
- [Locating and Controlling Implicit Personalization in Large Language Models](https://mlsanity.com/paper/2608.11735/) (2026)
- [Detection, Classification, and Mitigation of Gender Bias in Large Language Models](https://mlsanity.com/paper/2506.12527/) (2025)
- [BiasGym: A Simple and Generalizable Framework for Analyzing and Removing Biases through Injection](https://mlsanity.com/paper/2508.08855/) (2025)
- [Tracing the Latent Threads: A Mechanistic Study of How LLMs Represent and Operationalize Race and Ethnicity Cues](https://mlsanity.com/paper/2601.12868/) (2026)
- [Faithful-Patchscopes: Understanding and Mitigating Model Bias in Hidden Representations Explanation of Large Language Models](https://mlsanity.com/paper/2602.00300/) (2026)
- [Routing Sensitivity Without Controllability: A Diagnostic Study of Fairness in MoE Language Models](https://mlsanity.com/paper/2603.27141/) (2026)
- [STEREODISCO: Discovering Stereotypicality in LLMs](https://mlsanity.com/paper/2607.27824/) (2026)
- [GAPS: Dimension-Level Gates for Conditional Activation Steering](https://mlsanity.com/paper/2609.01878/) (2026)
- [Debiasing LLMs by Masking Unfairness-Driving Attention Heads](https://mlsanity.com/paper/2510.10142/) (2025)
- [Knowing Bias, Doing Better: Mitigating Social Bias in LLMs via Know-Bias Neuron Enhancement](https://mlsanity.com/paper/2601.21864/) (2026)
- [Pref-CTRL: Preference Driven LLM Alignment using Representation Editing](https://mlsanity.com/paper/2604.23543/) (2026)
- [Steered Generation via Gradient-Based Optimization on Sparse Query Features](https://mlsanity.com/paper/2605.23040/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are prone to capturing biases from training corpus, leading to potential negative social impacts. Existing prompt-based debiasing methods exhibit instability due to their sensitivity to prompt changes, while fine-tuning-based techniques incur substantial computational overhead and catastrophic forgetting. In this paper, we propose FairSteer, a novel inference-time debiasing framework without requiring customized prompt design or model retraining. Motivated by the linear representation hypothesis, our preliminary investigation demonstrates that fairness-related features can be encoded into separable directions in the hidden activation space. FairSteer operates in three steps: biased activation detection, debiasing steering vector (DSV) computation, and dynamic activation steering. Specifically, it first trains a lightweight linear classifier to detect bias signatures in activations, and then computes DSVs as intervention directions derived from small contrastive prompt pairs. Subsequently, it performs debiasing by adjusting activations with DSVs in the inference stage. Comprehensive evaluation with six LLMs demonstrates the superiority of FairSteer across question-answering, counterfactual input evaluation and open-ended text generation tasks. Code will be released.