On Temperature-Constrained Non-Deterministic Machine Translation: Potential and Evaluation

TL;DR

This paper explores temperature-constrained non-deterministic machine translation, revealing its potential to address multimodality issues, while also highlighting new evaluation challenges and proposing a robust selection strategy.

Contribution

It systematically evaluates ND-MT, identifies its advantages and challenges, and introduces ExpectoSample for reliable system evaluation and selection.

Findings

ND-MT can better address multimodality in translation.

Evaluation metrics are inconsistent for ND-MT, affected by the Buckets Effect.

ExpectoSample improves robustness in ND-MT system selection.

Abstract

In recent years, the non-deterministic properties of language models have garnered considerable attention and have shown a significant influence on real-world applications. However, such properties remain under-explored in machine translation (MT), a complex, non-deterministic NLP task. In this study, we systematically evaluate modern MT systems and identify temperature-constrained Non-Deterministic MT (ND-MT) as a distinct phenomenon. Additionally, we demonstrate that ND-MT exhibits significant potential in addressing the multimodality issue that has long challenged MT research and provides higher-quality candidates than Deterministic MT (D-MT) under temperature constraints. However, ND-MT introduces new challenges in evaluating system performance. Specifically, the evaluation framework designed for D-MT fails to yield consistent evaluation results when applied to ND-MT. We further investigate this emerging challenge by evaluating state-of-the-art ND-MT systems using both lexical-based and semantic-based metrics at varying sampling sizes. The results reveal a Buckets Effect across these systems: the ranking of ND-MT systems is dominated by the worst-quality candidate translation, as shown by automatic evaluation metrics. To mitigate this issue, we propose ExpectoSample, a strategy that first identifies reliable metrics and then enables robust ND-MT system selection for real-world.