An Interdisciplinary Approach to Human-Centered Machine Translation

2025-06-16

AI Summary

Machine translation (MT) is widely deployed in settings without professional translators, yet non-expert users struggle to assess its reliability, revealing a critical misalignment between system design and real-world usage needs. Method: This paper proposes a human-centered MT design paradigm, systematically integrating translation studies and human-computer interaction (HCI) theory for the first time—moving beyond BLEU-centric automatic evaluation toward a multidimensional framework grounded in users’ goals, capabilities, and contextual constraints. Through interdisciplinary literature review, situated needs analysis, human-centered evaluation modeling, and empirical studies of translation behavior, we develop a scalable set of human-centered MT design principles. Contribution/Results: The work establishes a foundational theoretical and practical framework for developing next-generation MT tools that are trustworthy, interpretable, and contextually adaptive—bridging the gap between technical performance and user-centered usability.

Technology Category

• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Multiagent Systems: Mechanism Design
• Humans and AI: Intelligent User Interfaces

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Machine Translation (MT) tools are widely used today, often in contexts where professional translators are not present. Despite progress in MT technology, a gap persists between system development and real-world usage, particularly for non-expert users who may struggle to assess translation reliability. This paper advocates for a human-centered approach to MT, emphasizing the alignment of system design with diverse communicative goals and contexts of use. We survey the literature in Translation Studies and Human-Computer Interaction to recontextualize MT evaluation and design to address the diverse real-world scenarios in which MT is used today.

Problem

• Bridging gap between MT development and real-world usage
• Improving translation reliability assessment for non-expert users
• Aligning MT design with diverse communicative goals

Innovation

• Human-centered Machine Translation approach
• Aligns system design with diverse goals
• Recontextualizes MT evaluation from literature