Skip to main navigation Skip to search Skip to main content

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

  • Stevens Institute of Technology

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

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.

Original languageEnglish
Title of host publicationMachine Learning and Principles and Practice of Knowledge Discovery in Databases - International Workshops of ECML PKDD 2025, Revised Selected Papers
EditorsIrena Koprinska, João Mendes-Moreira, Paula Branco
Pages277-293
Number of pages17
DOIs
StatePublished - 2026
EventEuropean Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2025 - Porto, Portugal
Duration: 15 Sep 202519 Sep 2025

Publication series

NameCommunications in Computer and Information Science
Volume2839 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

ConferenceEuropean Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2025
Country/TerritoryPortugal
CityPorto
Period15/09/2519/09/25

Keywords

  • Fairness in recommendations
  • Large language models (LLM)
  • LLM-based Recommendation systems

Fingerprint

Dive into the research topics of 'Lightweight Fairness for LLM-Based Recommendations via Kernelized Projection and Gated Adapters'. Together they form a unique fingerprint.

Cite this