Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper addresses the challenge of fairness in personalized text generation by framing it as a constrained multi-objective alignment problem, aiming to reduce demographic disparities while maintaining personalization fidelity. The authors introduce a Pareto-guided teacher alignment framework that integrates various techniques, including revision-based candidate generation and Pareto-style candidate selection, to evaluate its effectiveness on climate change and vaccination persuasion tasks. Results reveal that no single alignment strategy excels across all objectives, highlighting the importance of bounded-regression and multi-audit model selection for achieving fairness-sensitive personalized generation.
Fairness in personalized text generation is not a one-size-fits-all solution—different strategies yield trade-offs across demographic disparities and personalization fidelity.
Personalized persuasive text generation can improve relevance and engagement, but demographic conditioning may also introduce unequal framing across groups. We study fairness mitigation in personalized generation as a constrained multi-objective alignment problem: reduce demographic disparities while preserving personalization fidelity. We propose a Pareto-guided teacher alignment framework that combines revision-based candidate generation, pair-aware feasibility gating, Pareto-style candidate selection, and optional preference optimization through supervised fine-tuning and direct preference optimization. We evaluate the framework on climate change and vaccination persuasion tasks using a controlled context-rich demographic grid with matched gender and age pairs and a unified five-audit evaluation suite spanning persuasion bias, formality disparity, emotional framing disparity, lexical association disparity, and personalization fidelity. Across both domains and cross-family transfer settings, no single alignment strategy dominates all objectives simultaneously. Instead, methods occupy different regions of a fairness-personalization Pareto frontier: some achieve stronger disparity reductions, while others better preserve personalization or demographic stability. Our results show that fairness mitigation effects are objective-dependent and transfer inconsistently across domains and model families, motivating bounded-regression, multi-audit model selection over single-metric optimization for fairness-sensitive personalized generation.