When to Ask a Question: Understanding Communication Strategies in Generative AI Tools

TL;DR

This paper models user-LLM interactions to optimize information elicitation, balancing user burden and preference diversity, aiming to reduce bias and improve fairness in generative AI tools.

Contribution

It introduces a theoretical framework for balancing inference and elicitation in generative AI, supported by empirical validation.

Findings

Optimal information solicitation reduces bias in preference inference.

Elicitation strategies improve fairness without excessive user burden.

Model predictions align with empirical observations.

Abstract

Generative AI models differ from traditional machine learning tools in that they allow users to provide as much or as little information as they choose in their inputs. This flexibility often leads users to omit certain details, relying on the models to infer and fill in under-specified information based on distributional knowledge of user preferences. Such inferences may privilege majority viewpoints and disadvantage users with atypical preferences, raising concerns about fairness. Unlike more traditional recommender systems, LLMs can explicitly solicit more information from users through natural language. However, while directly eliciting user preferences could increase personalization and mitigate inequality, excessive querying places a burden on users who value efficiency. We develop a stylized model of user-LLM interaction and develop an objective that captures tradeoff between user burden and preference representation. Building on the observation that individual preferences are often correlated, we analyze how AI systems should balance inference and elicitation, characterizing the optimal amount of information to solicit before content generation. Ultimately, we show that information elicitation can mitigate the systematic biases of preference inference, enabling the design of generative tools that better incorporate diverse user perspectives while maintaining efficiency. We complement this theoretical analysis with an empirical evaluation illustrating the model's predictions and exploring their practical implications.