# Lightweight Fairness for LLM-Based Recommendations via Kernelized Projection and Gated Adapters

arXiv: https://arxiv.org/abs/2603.23780 · HTML: https://mlsanity.com/paper/2603.23780/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in MoE for recommender systems, one of 12 directions in Mixture-of-experts.
- Builds on 23 of its 34 references that are in the map.
- First posted on arXiv 24 Mar 2026.

## Placed in

- Mixture-of-experts / [MoE for recommender systems](https://mlsanity.com/moe/moe-for-recommender-systems/)

## Builds on (most cited first, 23 of 23)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Efficient Estimation of Word Representations in Vector Space](https://mlsanity.com/paper/1301.3781/) (2013)
- [Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings](https://mlsanity.com/paper/1607.06520/) (2016)
- [Counterfactual Fairness](https://mlsanity.com/paper/1703.06856/) (2017)
- [Bias and Fairness in Large Language Models: A Survey](https://mlsanity.com/paper/2309.00770/) (2023)
- [Null It Out: Guarding Protected Attributes by Iterative Nullspace Projection](https://mlsanity.com/paper/2004.07667/) (2020)
- [Recommendation as Language Processing (RLP): A Unified Pretrain, Personalized Prompt & Predict Paradigm (P5)](https://mlsanity.com/paper/2203.13366/) (2022)
- [Fairness in Machine Learning: A Survey](https://mlsanity.com/paper/2010.04053/) (2020)
- [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)
- [Chat-REC: Towards Interactive and Explainable LLMs-Augmented Recommender System](https://mlsanity.com/paper/2303.14524/) (2023)
- [Recommender Systems in the Era of Large Language Models (LLMs)](https://mlsanity.com/paper/2307.02046/) (2023)
- [Is ChatGPT Fair for Recommendation? Evaluating Fairness in Large Language Model Recommendation](https://mlsanity.com/paper/2305.07609/) (2023)
- [Linear Adversarial Concept Erasure](https://mlsanity.com/paper/2201.12091/) (2022)
- [CoLLM: Integrating Collaborative Embeddings into Large Language Models for Recommendation](https://mlsanity.com/paper/2310.19488/) (2023)
- [UP5: Unbiased Foundation Model for Fairness-aware Recommendation](https://mlsanity.com/paper/2305.12090/) (2023)
- [Item-side Fairness of Large Language Model-based Recommendation System](https://mlsanity.com/paper/2402.15215/) (2024)
- [CFaiRLLM: Consumer Fairness Evaluation in Large-Language Model Recommender System](https://mlsanity.com/paper/2403.05668/) (2024)
- [Selective Fairness in Recommendation via Prompts](https://mlsanity.com/paper/2205.04682/) (2022)
- [RecGPT: Generative Pre-training for Text-based Recommendation](https://mlsanity.com/paper/2405.12715/) (2024)
- [On the Importance of Gaussianizing Representations](https://mlsanity.com/paper/2505.00685/) (2025)
- [Oblivious Data for Fairness with Kernels](https://mlsanity.com/paper/2002.02901/) (2020)
- [Kernel Debiased Plug-in Estimation: Simultaneous, Automated Debiasing without Influence Functions for Many Target Parameters](https://mlsanity.com/paper/2306.08598/) (2023)
- [GATE X-E : A Challenge Set for Gender-Fair Translations from Weakly-Gendered Languages](https://mlsanity.com/paper/2402.14277/) (2024)

## Built on by (most cited first, 1 of 1)

- [Rethinking Fairness in LLM-Based Recommender Systems: A Survey](https://mlsanity.com/paper/2606.28340/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have introduced new capabilities to recommender systems, enabling dynamic, context-aware, and conversational recommendations. However, LLM-based recommender systems inherit and may amplify social biases embedded in their pre-training data, especially when demographic cues are present. Existing fairness solutions either require extra parameters fine-tuning, or suffer from optimization instability. We propose a lightweight and scalable bias mitigation method that combines a kernelized Iterative Null-space Projection (INLP) with a gated Mixture-of-Experts (MoE) adapter. Our approach estimates a closed-form projection that removes single or multiple sensitive attributes from LLM representations with no additional trainable parameters. To preserve task utility, we introduce a two-level MoE adapter that selectively restores useful signals without reintroducing bias. Experiments on two public datasets show that our method reduces attribute leakage across multiple protected variables while maintaining competitive recommendation accuracy.