Search papers, labs, and topics across Lattice.
This paper introduces RobotValues, a benchmark designed to evaluate household robots in scenarios where human values conflict, such as autonomy and efficiency. By generating 10,000 value-conflict scenarios through LLM-assisted methods, the authors assess how well robot planners can prioritize different human values beyond mere task completion. The evaluation reveals that existing models default to certain values like safety, often neglecting privacy considerations, and struggle to adapt their actions when instructed to prioritize conflicting values, making incorrect choices 80% of the time.
Robots often fail to adapt their actions to prioritize human values, making incorrect choices 80% of the time in value-conflicted scenarios.
While household robots are often evaluated based on task completion, everyday domestic environments involve value-conflicting situations in which robots are expected to choose actions that prioritize other values than task success, such as human autonomy, efficiency, or social appropriateness. Yet, there are no benchmarks for evaluating robots'value preferences in such scenarios. We introduce RobotValues, a benchmark to evaluate household robot planners in 10K value-conflict scenarios. Each instance consists of a realistic household image with multiple plausible robot actions that prioritize different human values. We construct RobotValues through LLM-assisted scenario generation, stakeholder-grounded value extraction, image generation and automatic quality control. Using RobotValues we evaluate VLMs used in robotics and find that models exhibit default value preferences, including safety and accommodation, while underselecting privacy-prioritizing actions. When the models are instructed to prioritize specific values that conflict with their own preferences, they often fail to override their default actions, choosing incorrect actions for 80% of the time. These findings suggest that household robot evaluation should measure not only task completion or safety compliance, but also whether robots can choose among plausible actions when human values conflict.
