ALIGN: Prompt-based Attribute Alignment for Reliable, Responsible, and Personalized LLM-based Decision-Making

Large language models (LLMs) are increasingly being used as decision aids. However, users have diverse values and preferences that can affect their decision-making, which requires novel methods for LLM alignment and personalization. Existing LLM comparison tools largely focus on benchmarking tasks, such as knowledge-based question answering. In contrast, our proposed ALIGN system focuses on dynamic personalization of LLM-based decision-makers through prompt-based alignment to a set of fine-grained attributes. Key features of our system include robust configuration management, structured output generation with reasoning, and several algorithm implementations with swappable LLM backbones, enabling different types of analyses.

3.1. Core Software Framework

The ALIGN system framework is an open source Python module that allows users to: (1) implement and configure ADMs, and (2) run ADMs through a series of questions via a dataset interface (see Figure 2). The dataset interface provides domain-specific information for a given scenario. In the medical triage domain, this may include a high-level text description of the situation, patient descriptions along with vital signs and injuries, as well as available treatment supplies. In the demographic attribute alignment domain, this is an open-ended survey question that evokes diverse views or opinions. Alignable ADMs also utilize an attribute alignment target to guide the decision-making process.