发表机构
Human Language Technologies Research Center; Faculty of Mathematics and Computer Science; University of Bucharest（人类语言技术研究中心; 数学与计算机科学学院; 布加勒斯特大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出结合LLM性别分类与Transformer翻译的混合流水线，引入新数据集，使英罗机器翻译在WinoMT等基准上性别准确率提升超40个百分点，为缓解该领域性别偏差提供新方法。
AI 中文摘要
机器翻译（MT）系统常无法正确翻译性别，尤其在从英语这种中性语言转换到罗马尼亚语这种有性别的目标语言时。这种偏差会导致译文默认使用阳性形式或强化性别刻板印象。我们提出一种混合流水线来缓解该问题，结合基于大语言模型（LLM）的性别分类与神经机器翻译（NMT）。我们的系统使用微调后的LLM检测英语句子中目标词的意图性别，并插入内联性别提示标签，这些带标签的句子随后被传入微调过的Transformer模型，以生成形态正确的罗马尼亚语译文。为支持该工作，我们引入三个用于性别消歧和翻译的新数据集。与基线MT系统相比，我们的方法在WinoMT和WinoGender基准上的性别准确率提升超过40个百分点。这是首个同时使用LLM推理和标签感知翻译来明确处理并评估英罗机器翻译中性别偏差的方法。
英文摘要
Machine translation (MT) systems often fail to correctly translate gender, especially when converting from a gender-neutral language like English to a gendered target language such as Romanian. This bias results in translations that default to masculine forms or reinforce gender stereotypes. We propose a hybrid pipeline to mitigate this issue by combining large language model (LLM)-based gender classification with neural machine translation (NMT). Our system uses a fine-tuned LLM to detect the intended gender of target words in English sentences and insert inline gender hint tags. These tagged sentences are then passed to a Transformer model fine-tuned to generate morphologically correct Romanian translations. To support this, we introduce three novel datasets for gender disambiguation and translation. Our approach improves gender accuracy on the WinoMT and WinoGender benchmarks by over 40 percentage points compared to a baseline MT system. This is the first method to explicitly address and evaluate gender bias in English-Romanian MT using both LLM inference and tag-aware translation.
DOI：10.1007/978-3-032-29532-3_11
