Distill: Uncovering the True Intent behind Human-Robot Communication

2026-05-13

AI Summary

This work addresses the challenge that natural language instructions are often ambiguous while end-user programming is overly specific, hindering robots from accurately capturing users’ true intent. To bridge this semantic gap, the paper introduces Distill, a novel method that systematically integrates three stages—simplification (removing redundant steps), generalization (abstracting the semantics of individual actions), and sequential relaxation (loosening ordering constraints)—to distill core task intent from initial user descriptions. Implemented via a web-based interface that combines task parsing, semantic generalization, and constraint relaxation algorithms, Distill significantly improves both the accuracy of intent understanding and the flexibility of task execution in human-robot interaction, as demonstrated through crowdsourced experiments.

Abstract

As robots become increasingly integrated into everyday environments, intuitive communication paradigms such as natural language and end-user programming have become indispensable for specifying autonomous robot behavior. However, these mechanisms are ineffective at fully capturing user intent: natural language is imprecise and ambiguous, whereas end-user programming can be overly specific. As a result, understanding what users truly mean when they interact with robots remains a central challenge for human-AI communication systems. To address this issue, we propose the Distill approach for human-robot communication interfaces. Given a task specification provided by the user, Distill (1) removes unnecessary steps; (2) generalizes the meaning behind individual steps; and (3) relaxes ordering constraints between steps. We implemented Distill on a web interface and, through a crowdsourcing study, demonstrated its ability to elicit and refine user intent from initial task specifications.

Problem

• human-robot communication
• user intent
• natural language ambiguity
• end-user programming
• intent understanding

Innovation

• Distill
• human-robot communication
• user intent
• task specification
• intent generalization