When Personalization Meets Reality: A Multi-Faceted Analysis of Personalized Preference Learning

• URL: http://arxiv.org/abs/2502.19158v1
• Date: Wed, 26 Feb 2025 14:14:58 GMT
• Title: When Personalization Meets Reality: A Multi-Faceted Analysis of Personalized Preference Learning
• Authors: Yijiang River Dong, Tiancheng Hu, Yinhong Liu, Ahmet Üstün, Nigel Collier,
• Abstract summary: Reinforcement Learning from Human Feedback (RLHF) typically assumes homogeneous preferences across users, overlooking diverse human values and minority viewpoints.<n>We present a multi-faceted evaluation framework that measures not only performance but also fairness, unintended effects, and adaptability across varying levels of preference divergence.<n>These findings highlight the critical need for holistic evaluation approaches to advance the development of more effective and inclusive preference learning systems.
• Score: 23.557084253364174
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: While Reinforcement Learning from Human Feedback (RLHF) is widely used to align Large Language Models (LLMs) with human preferences, it typically assumes homogeneous preferences across users, overlooking diverse human values and minority viewpoints. Although personalized preference learning addresses this by tailoring separate preferences for individual users, the field lacks standardized methods to assess its effectiveness. We present a multi-faceted evaluation framework that measures not only performance but also fairness, unintended effects, and adaptability across varying levels of preference divergence. Through extensive experiments comparing eight personalization methods across three preference datasets, we demonstrate that performance differences between methods could reach 36% when users strongly disagree, and personalization can introduce up to 20% safety misalignment. These findings highlight the critical need for holistic evaluation approaches to advance the development of more effective and inclusive preference learning systems.

Related papers

• Uncertain Multi-Objective Recommendation via Orthogonal Meta-Learning Enhanced Bayesian Optimization [30.031396809114625]
  We introduce a novel framework that categorizes RS autonomy into five distinct levels, ranging from basic rule-based accuracy-driven systems to behavior-aware, uncertain multi-objective RSs. We propose an approach that dynamically identifies and optimize multiple objectives based on individual user preferences, fostering more ethical and intelligent user-centric recommendations.
  arXiv  Detail & Related papers  (2025-02-18T08:10:09Z)
• Few-shot Steerable Alignment: Adapting Rewards and LLM Policies with Neural Processes [50.544186914115045]
  Large language models (LLMs) are increasingly embedded in everyday applications. Ensuring their alignment with the diverse preferences of individual users has become a critical challenge. We present a novel framework for few-shot steerable alignment.
  arXiv  Detail & Related papers  (2024-12-18T16:14:59Z)
• ComPO: Community Preferences for Language Model Personalization [122.54846260663922]
  ComPO is a method to personalize preference optimization in language models. We collect and release ComPRed, a question answering dataset with community-level preferences from Reddit.
  arXiv  Detail & Related papers  (2024-10-21T14:02:40Z)
• Aligning LLMs with Individual Preferences via Interaction [51.72200436159636]
  We train large language models (LLMs) that can ''interact to align''<n>We develop a multi-turn preference dataset containing 3K+ multi-turn conversations in tree structures.<n>For evaluation, we establish the ALOE benchmark, consisting of 100 carefully selected examples and well-designed metrics to measure the customized alignment performance during conversations.
  arXiv  Detail & Related papers  (2024-10-04T17:48:29Z)
• Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learning [12.742158403867002]
  Reinforcement Learning from Human Feedback is a powerful paradigm for aligning foundation models to human values and preferences. Current RLHF techniques cannot account for the naturally occurring differences in individual human preferences across a diverse population. We develop a class of multimodal RLHF methods to address the need for pluralistic alignment.
  arXiv  Detail & Related papers  (2024-08-19T15:18:30Z)
• Controllable Preference Optimization: Toward Controllable Multi-Objective Alignment [103.12563033438715]
  Alignment in artificial intelligence pursues consistency between model responses and human preferences as well as values. Existing alignment techniques are mostly unidirectional, leading to suboptimal trade-offs and poor flexibility over various objectives. We introduce controllable preference optimization (CPO), which explicitly specifies preference scores for different objectives.
  arXiv  Detail & Related papers  (2024-02-29T12:12:30Z)
• MaxMin-RLHF: Alignment with Diverse Human Preferences [101.57443597426374]
  Reinforcement Learning from Human Feedback (RLHF) aligns language models to human preferences by employing a singular reward model derived from preference data.<n>We learn a mixture of preference distributions via an expectation-maximization algorithm to better represent diverse human preferences.<n>Our algorithm achieves an average improvement of more than 16% in win-rates over conventional RLHF algorithms.
  arXiv  Detail & Related papers  (2024-02-14T03:56:27Z)
• Which Prompts Make The Difference? Data Prioritization For Efficient Human LLM Evaluation [9.452326973655445]
  We find that metric-based methods enhance the efficiency of human evaluations by minimizing the number of required annotations. We show that our method is effective across widely used model families, reducing instances of indecisive (or "tie") outcomes by up to 54%. This potential reduction in required human effort positions our approach as a valuable strategy in future large language model evaluations.
  arXiv  Detail & Related papers  (2023-10-22T21:48:51Z)
• Everyone Deserves A Reward: Learning Customized Human Preferences [25.28261194665836]
  Reward models (RMs) are essential for aligning large language models with human preferences to improve interaction quality. We propose a three-stage customized RM learning scheme, then empirically verify its effectiveness on both general preference datasets and our DSP set. We find several ways to better preserve the general preferring ability while training the customized RMs.
  arXiv  Detail & Related papers  (2023-09-06T16:03:59Z)
• MetaAge: Meta-Learning Personalized Age Estimators [94.73054410570037]
  We propose a meta-learning method named MetaAge for age estimation. Specifically, we introduce a personalized estimator meta-learner, which takes identity features as the input and outputs the parameters of customized estimators. In this way, our method learns the meta knowledge without the above requirements and seamlessly transfers the learned meta knowledge to the test set.
  arXiv  Detail & Related papers  (2022-07-12T03:53:42Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.