Should I Share this Translation? Evaluating Quality Feedback for User Reliance on Machine Translation

TL;DR

This study evaluates various feedback methods to help users better assess machine translation quality, finding that implicit feedback, especially QA tables, enhances decision accuracy and user trust more effectively than explicit feedback.

Contribution

It introduces and compares four types of quality feedback for MT, highlighting the superior effectiveness of implicit feedback like QA tables in improving user reliance and perception.

Findings

Implicit feedback, especially QA tables, significantly improves decision accuracy.

All feedback types except error highlights enhance appropriate reliance.

Implicit feedback receives higher user ratings for helpfulness and trust.

Abstract

As people increasingly use AI systems in work and daily life, feedback mechanisms that help them use AI responsibly are urgently needed, particularly in settings where users are not equipped to assess the quality of AI predictions. We study a realistic Machine Translation (MT) scenario where monolingual users decide whether to share an MT output, first without and then with quality feedback. We compare four types of quality feedback: explicit feedback that directly give users an assessment of translation quality using (1) error highlights and (2) LLM explanations, and implicit feedback that helps users compare MT inputs and outputs through (3) backtranslation and (4) question-answer (QA) tables. We find that all feedback types, except error highlights, significantly improve both decision accuracy and appropriate reliance. Notably, implicit feedback, especially QA tables, yields significantly greater gains than explicit feedback in terms of decision accuracy, appropriate reliance, and user perceptions, receiving the highest ratings for helpfulness and trust, and the lowest for mental burden.