ValueCompass: A Framework for Measuring Contextual Value Alignment Between Human and LLMs

2024-09-15

AI Summary

This study quantifies the alignment between large language models (LLMs) and human values. To address the lack of a systematic, cross-context value framework, we propose ValueCompass—a psychologically grounded value taxonomy that formally defines measurable, domain-agnostic value dimensions across four high-stakes scenarios: collaborative writing, education, public affairs, and healthcare. Our methodology integrates literature synthesis, multi-scenario empirical surveys of human value preferences, and behavioral evaluation of LLMs under context-specific prompts. Results reveal significant structural misalignment—particularly on core values such as “national security”—and demonstrate strong contextual sensitivity in human value judgments. Our contributions are threefold: (1) the first reusable, cross-domain benchmark for value alignment assessment; (2) empirical validation that contextual dynamics critically influence AI alignment outcomes; and (3) theoretically grounded, actionable guidelines for designing and evaluating trustworthy AI systems.

Technology Category

• Humans and AI: Learning Human Values and Preferences
• Philosophy and Ethics of AI: Morality & Value-based AI
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

As AI systems become more advanced, ensuring their alignment with a diverse range of individuals and societal values becomes increasingly critical. But how can we capture fundamental human values and assess the degree to which AI systems align with them? We introduce ValueCompass, a framework of fundamental values, grounded in psychological theory and a systematic review, to identify and evaluate human-AI alignment. We apply ValueCompass to measure the value alignment of humans and large language models (LLMs) across four real-world scenarios: collaborative writing, education, public sectors, and healthcare. Our findings reveal concerning misalignments between humans and LLMs, such as humans frequently endorse values like"National Security"which were largely rejected by LLMs. We also observe that values differ across scenarios, highlighting the need for context-aware AI alignment strategies. This work provides valuable insights into the design space of human-AI alignment, laying the foundations for developing AI systems that responsibly reflect societal values and ethics.

Problem

• Measure human-AI value alignment across diverse contexts
• Identify concerning misalignments between humans and LLMs
• Develop context-aware strategies for ethical AI design

Innovation

• Framework for human-AI value alignment measurement
• Context-aware AI alignment strategies
• Psychological theory-based value evaluation