QUEST：用于机器翻译的质量感知Metropolis-Hastings采样
QUEST: Quality-Aware Metropolis-Hastings Sampling for Machine Translation
AI总结：
本文提出QUEST方法，利用Metropolis-Hastings采样从Gibbs分布的高密度区域生成高质量且多样化的机器翻译，避免单一重排序带来的指标作弊问题。
摘要：
机器翻译（MT）的一个重要挑战是生成高质量和多样化的翻译。先前的工作表明，MT模型估计的似然性与翻译质量的相关性很差。相比之下，质量评估指标（如COMET或BLEURT）与人类判断表现出高度相关性，这促使它们被用作重排序器（如质量感知和最小贝叶斯风险解码）。然而，依赖具有高估计质量的单一翻译会增加“钻指标空子”的几率。在本文中，我们解决了采样一组高质量和多样化翻译的问题。我们提供了一种简单有效的方法，通过将噪声质量估计用作Gibbs分布的能量函数，来避免过度依赖这些估计。我们不是在分布中寻找众数，而是通过Metropolis-Hastings算法（一种简单的马尔可夫链蒙特卡洛方法）从高密度区域生成多个样本。结果表明，我们提出的方法在两个强大的仅解码器LLM（Alma-7b, Tower-7b）下，在多个语言对（英语$\leftrightarrow${德语，俄语}）上均能产生高质量和多样化的输出。
英文摘要：
An important challenge in machine translation (MT) is to generate high-quality and diverse translations. Prior work has shown that the estimated likelihood from the MT model correlates poorly with translation quality. In contrast, quality evaluation metrics (such as COMET or BLEURT) exhibit high correlations with human judgments, which has motivated their use as rerankers (such as quality-aware and minimum Bayes risk decoding). However, relying on a single translation with high estimated quality increases the chances of "gaming the metric''. In this paper, we address the problem of sampling a set of high-quality and diverse translations. We provide a simple and effective way to avoid over-reliance on noisy quality estimates by using them as the energy function of a Gibbs distribution. Instead of looking for a mode in the distribution, we generate multiple samples from high-density areas through the Metropolis-Hastings algorithm, a simple Markov chain Monte Carlo approach. The results show that our proposed method leads to high-quality and diverse outputs across multiple language pairs (English$\leftrightarrow${German, Russian}) with two strong decoder-only LLMs (Alma-7b, Tower-7b).
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
