ValuePilot: A Two-Phase Framework for Value-Driven Decision-Making

2025-12-09

AI Summary

This paper addresses the challenge of enabling AI agents to consistently and interpretable adapt to individuals’ deep-value preferences. Methodologically: (1) we introduce a human-in-the-loop value annotation toolkit (DGT) to efficiently generate fine-grained, individual-level value preference datasets; (2) we propose a value-aware decision-making module (DMM) that embeds individual values into action evaluation and enhances robustness via cross-situational generalization training. Our key contribution is the first paradigm shift from collective value alignment to individual value alignment, enabling dynamic adaptation and behavior interpretability in open-ended scenarios. Experimental results demonstrate that DMM significantly outperforms state-of-the-art foundation models—including GPT-4o, Claude-3.5-Sonnet, Gemini-2.0-Flash, and Llama-3.1-70B—in aligning with human value choices on unseen scenarios.

Technology Category

• Humans and AI: Learning Human Values and Preferences
• Multiagent Systems: Adversarial Agents
• Philosophy and Ethics of AI: Morality & Value-based AI

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Personalized decision-making is essential for human-AI interaction, enabling AI agents to act in alignment with individual users' value preferences. As AI systems expand into real-world applications, adapting to personalized values beyond task completion or collective alignment has become a critical challenge. We address this by proposing a value-driven approach to personalized decision-making. Human values serve as stable, transferable signals that support consistent and generalizable behavior across contexts. Compared to task-oriented paradigms driven by external rewards and incentives, value-driven decision-making enhances interpretability and enables agents to act appropriately even in novel scenarios. We introduce ValuePilot, a two-phase framework consisting of a dataset generation toolkit (DGT) and a decision-making module (DMM). DGT constructs diverse, value-annotated scenarios from a human-LLM collaborative pipeline. DMM learns to evaluate actions based on personal value preferences, enabling context-sensitive, individualized decisions. When evaluated on previously unseen scenarios, DMM outperforms strong LLM baselines, including GPT-5, Claude-Sonnet-4, Gemini-2-flash, and Llama-3.1-70b, in aligning with human action choices. Our results demonstrate that value-driven decision-making is an effective and extensible engineering pathway toward building interpretable, personalized AI agents.

Problem

• Develops a framework for personalized AI decision-making based on individual human values.
• Generates diverse value-annotated scenarios to train context-sensitive decision models.
• Enhances interpretability and alignment with human preferences in novel situations.

Innovation

• Two-phase framework for value-driven personalized decision-making
• Dataset generation toolkit creates value-annotated scenarios collaboratively
• Decision-making module learns to evaluate actions based on personal values