Interactive system helps clarify vague user attribute descriptions
Let Me Clarify: Interactive Modeling of Soft Attributes
Human-Computer Interaction
Summary
People often describe preferences using soft words like “interesting,” which can mean different things to different people and are hard for computers to understand. The authors define how to measure the flexibility of such descriptions and created a system called AttriFy to turn these vague ideas into clear, programmable functions. This system helps users refine their preferences through an interactive process, making it easier to see, accept, or change how their words are interpreted. In tests with 32 participants, the approach improved how people built understandable preferences compared to a baseline using large language models.
What this means in practice
- •For ux designers: Create interactive tools that help users define and clarify subjective preferences for personalized content or product recommendations.
- •For customer support teams: Develop chatbot assistants that clarify vague customer inputs into precise queries for better automated responses.
Authors
Steffen Holter, Rita Sevastjanova, Michaela Benk, Mennatallah El-Assady
Abstract
People commonly express preferences through soft attributes such as is interesting,'' which are difficult to operationalize for computational preference modeling. Latent representations can obscure how such attributes are interpreted, limiting transparency and user control. We formalize softness as the diversity of valid interpretations a description permits and provide empirical support for this characterization. Our proposed measure estimates softness to guide clarification and determine when to generate computational approximations. This is instantiated in AttriFy, a mixed-initiative system that translates descriptions into executable Python functions over text corpora. An interactive dashboard connects the dialogue to proposed functions and computed values, enabling users to inspect, accept, reject, and refine interpretations. An exploratory within-subjects study with 32 participants compares our softness-guided clarification approach with an LLM baseline in supporting interpretable attribute construction. This work positions softness as a basis for constructing explicit, user-steerable attribute dimensions for downstream preference elicitation and decision-making.