# BiasFilter: An Inference-Time Debiasing Framework for Large Language Models

arXiv: https://arxiv.org/abs/2505.23829 · HTML: https://mlsanity.com/paper/2505.23829/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 35 of its 60 references that are in the map.
- First posted on arXiv 28 May 2025.

## Builds on (most cited first, 30 of 35)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [A Diversity-Promoting Objective Function for Neural Conversation Models](https://mlsanity.com/paper/1510.03055/) (2015)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [From r to Q^*: Your Language Model is Secretly a Q-Function](https://mlsanity.com/paper/2404.12358/) (2024)
- [Large image datasets: A pyrrhic win for computer vision?](https://mlsanity.com/paper/2006.16923/) (2020)
- [OpenAGI: When LLM Meets Domain Experts](https://mlsanity.com/paper/2304.04370/) (2023)
- [Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventions](https://mlsanity.com/paper/2306.04140/) (2023)
- [Cognitive Bias in Decision-Making with LLMs](https://mlsanity.com/paper/2403.00811/) (2024)
- [ARGS: Alignment as Reward-Guided Search](https://mlsanity.com/paper/2402.01694/) (2024)
- [Bias in Large Language Models: Origin, Evaluation, and Mitigation](https://mlsanity.com/paper/2411.10915/) (2024)
- [Fast Model Debias with Machine Unlearning](https://mlsanity.com/paper/2310.12560/) (2023)
- [Self-Contrast: Better Reflection Through Inconsistent Solving Perspectives](https://mlsanity.com/paper/2401.02009/) (2024)
- [Exploring the Limits of Domain-Adaptive Training for Detoxifying Large-Scale Language Models](https://mlsanity.com/paper/2202.04173/) (2022)
- [TreeBoN: Enhancing Inference-Time Alignment with Speculative Tree-Search and Best-of-N Sampling](https://mlsanity.com/paper/2410.16033/) (2024)
- [LMFlow: An Extensible Toolkit for Finetuning and Inference of Large Foundation Models](https://mlsanity.com/paper/2306.12420/) (2023)
- [PAD: Personalized Alignment of LLMs at Decoding-Time](https://mlsanity.com/paper/2410.04070/) (2024)
- [Self-Debiasing Large Language Models: Zero-Shot Recognition and Reduction of Stereotypes](https://mlsanity.com/paper/2402.01981/) (2024)
- [Bias Testing and Mitigation in LLM-based Code Generation](https://mlsanity.com/paper/2309.14345/) (2023)
- [Evaluating Gender Bias in Large Language Models via Chain-of-Thought Prompting](https://mlsanity.com/paper/2401.15585/) (2024)
- [Critic-Guided Decoding for Controlled Text Generation](https://mlsanity.com/paper/2212.10938/) (2022)
- [Using In-Context Learning to Improve Dialogue Safety](https://mlsanity.com/paper/2302.00871/) (2023)
- [FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering](https://mlsanity.com/paper/2504.14492/) (2025)
- [FairMT-Bench: Benchmarking Fairness for Multi-turn Dialogue in Conversational LLMs](https://mlsanity.com/paper/2410.19317/) (2024)
- [Causal Prompting: Debiasing Large Language Model Prompting based on Front-Door Adjustment](https://mlsanity.com/paper/2403.02738/) (2024)
- [First the worst: Finding better gender translations during beam search](https://mlsanity.com/paper/2104.07429/) (2021)
- [Learnable Privacy Neurons Localization in Language Models](https://mlsanity.com/paper/2405.10989/) (2024)
- [BOLT: Fast Energy-based Controlled Text Generation with Tunable Biases](https://mlsanity.com/paper/2305.12018/) (2023)

## Built on by (most cited first, 6 of 6)

- [Self-Debias: Self-correcting for Debiasing Large Language Models](https://mlsanity.com/paper/2604.08243/) (2026)
- [Detection, Classification, and Mitigation of Gender Bias in Large Language Models](https://mlsanity.com/paper/2506.12527/) (2025)
- [Counting Worlds Branching Time Semantics for post-hoc Bias Mitigation in generative AI](https://mlsanity.com/paper/2604.19431/) (2026)
- [No-Free-Fairness: Fundamental Limits and Trade-offs in Learning Systems](https://mlsanity.com/paper/2606.17810/) (2026)
- [The Safeguard Worked. Is the LLM System Safer?](https://mlsanity.com/paper/2609.00519/) (2026)
- [Acmite: Mitigating Gender Bias in LLMs through Concept-Guided Mutual Information](https://mlsanity.com/paper/2610.01696/) (2026)

## Abstract (arXiv)

Mitigating social bias in large language models (LLMs) has become an increasingly important research objective. However, existing debiasing methods often incur high human and computational costs, exhibit limited effectiveness, and struggle to scale to larger models and open-ended generation tasks. To address these limitations, this paper proposes BiasFilter, a model-agnostic, inference-time debiasing framework that integrates seamlessly with both open-source and API-based LLMs. Instead of relying on retraining with balanced data or modifying model parameters, BiasFilter enforces fairness by filtering generation outputs in real time. Specifically, it periodically evaluates intermediate outputs every few tokens, maintains an active set of candidate continuations, and incrementally completes generation by discarding low-reward segments based on a fairness reward signal. To support this process, we construct a fairness preference dataset and train an implicit reward model to assess token-level fairness in generated responses. Extensive experiments demonstrate that BiasFilter effectively mitigates social bias across a range of LLMs while preserving overall generation quality.