paper-with-me

홈 › Papers

Translationese as a Language in "Multilingual" NMT

2019-11-10 · Parker Riley, Isaac Caswell, Markus Freitag, David Grangier

Machine translation has an undesirable propensity to produce "translationese" artifacts, which can lead to higher BLEU scores while being liked less by human raters. Motivated by this, we model translationese and original (i.e. natural) text as separate languages in a multilingual model, and pose the question: can we perform zero-shot translation between original source text and original target text? There is no data with original source and original target, so we train sentence-level classifiers to distinguish translationese from original target text, and use this classifier to tag the training data for an NMT model. Using this technique we bias the model to produce more natural outputs at test time, yielding gains in human evaluation scores on both accuracy and fluency. Additionally, we demonstrate that it is possible to bias the model to produce translationese and game the BLEU score, increasing it while decreasing human-rated quality. We analyze these models using metrics to measure the degree of translationese in the output, and present an analysis of the capriciousness of heuristically-based train-data tagging.

📄 PDF Abstract BibTeX arXiv:1911.03823

Code (0)

등록된 구현이 없습니다.

Tasks

Machine TranslationNMTSentenceTAGTranslation

Methods 이 논문이 사용한 방법론

Test 설명 없음

Similar Papers 제목 키워드 기반

An Investigation of Translationese in the Generations of Multilingual Large Language Models

2026-08-18 · Maria Valentini, Téa Wright, Julisa Granados, Eliana Colunga 외 arxiv

Text which has been translated from another language tends to carry with it evidence of translation$\unicode{x2014}$hence, it is often referred to as $\textit{translationese}$. Multilingual large language models (MLLMs) …

Translate-Train Embracing Translationese Artifacts

2022-05-01 · ACL 2022 5 · Sicheng Yu, Qianru Sun, Hao Zhang, Jing Jiang

Translate-train is a general training approach to multilingual tasks. The key idea is to use the translator of the target language to generate training data to mitigate the gap between the source and target languages. Ho…

Mitigating Translationese Bias in Multilingual LLM-as-a-Judge via Disentangled Information Bottleneck

2026-03-11 · Hongbin Zhang, Kehai Chen, Xuefen Bai, Youcheng Pan 외 arxiv

Large language models (LLMs) have become a standard for multilingual evaluation, yet they exhibit a severe systematic translationese bias. In this paper, translationese bias is characterized as LLMs systematically favori…

Translationese as a Language in ``Multilingual'' NMT

2020-07-01 · ACL 2020 6 · Parker Riley, Isaac Caswell, Markus Freitag, David Grangier

Machine translation has an undesirable propensity to produce {``}translationese{''} artifacts, which can lead to higher BLEU scores while being liked less by human raters. Motivated by this, we model translationese and o…

Machine TranslationNMTSentenceTAG+1

A Dataset for Probing Translationese Preferences in English-to-Swedish Translation

2026-03-09 · Jenny Kunz, Anja Jarochenko, Marcel Bollmann arxiv

Translations often carry traces of the source language, a phenomenon known as translationese. We introduce the first freely available English-to-Swedish dataset contrasting translationese sentences with idiomatic alterna…