arXiv:2606.24644v1 Announce Type: new Abstract: Millions of people use machine translation (MT) tools daily, yet little is known about their perception of what systems can and cannot do. This paper studies users' mental models of speech translation systems through a new framework based on cross-lingual question answering, where users either accept MT output or request professional re-translation to answer questions based on the information presented in a foreign language. By analyzing user behavior and accuracy trends across varying translation qualities, we examine to what extent they can pre
The proliferation of machine translation tools and the increasing integration of AI into daily human-computer interaction necessitate a deeper understanding of user perception and mental models.
Understanding how users perceive and interact with AI-powered translation tools is crucial for improving user experience, enhancing trust, and guiding future AI development towards more intuitive and effective collaboration.
This research provides a new framework for evaluating user understanding of AI translation capabilities, moving beyond simple accuracy metrics to psychological and behavioral insights.
- · AI/ML researchers
- · Machine translation developers
- · Companies investing in human-AI collaboration tools
- · Users of translation tools
- · Developers ignoring user mental models
- · Generic, one-size-fits-all AI models
Improved design and user interfaces for AI-driven language tools based on better understanding of user expectations.
Increased adoption and reliance on AI translation as user trust and satisfaction grow with more transparent and predictable system behavior.
Potential for AI systems to dynamically adapt their output or confidence levels based on real-time assessment of individual user mental models and needs.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL