发表机构
Nara Institute of Science and Technology (NAIST); The University of Tokyo（奈良科学技术研究所（NAIST）; 东京大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究针对最小贝叶斯风险解码的度量过拟合问题，提出SVD-MBR方法，通过奇异值分解解耦真实共识与度量噪声，在广义度量上取得显著性能提升，且去噪效果具度量依赖性。
AI 中文摘要
最小贝叶斯风险（MBR）解码通过选择在采样伪参考上最大化效用度量的假设，实现高质量文本生成，但它极易受到度量过拟合的影响：它会以牺牲其他未优化的评估度量为代价，不规则地放大所选的效用度量。为缓解这一问题，我们提出SVD-MBR，将成对效用矩阵视为含噪信息信号，通过奇异值分解（SVD）计算低秩近似并仅保留前k个分量，有效将真实共识与度量噪声解耦。实验表明，SVD-MBR成功正则化解码，在一系列广义度量上取得显著提升。此外，我们发现这种去噪具有度量依赖性：神经度量编码适合SVD的鲁棒低秩共识理想，而表面级度量难以将信号与度量噪声分离。
英文摘要
Minimum Bayes Risk (MBR) decoding enables high-quality text generation by selecting the hypothesis that maximizes a utility metric over sampled pseudo-references. However, it is highly susceptible to metric overfitting: it can irregularly inflate the chosen utility metric at the direct expense of other unoptimized evaluation metrics. To mitigate this, we introduce SVD-MBR, which frames the pairwise utility matrix as a noisy information signal. By computing a low-rank approximation via Singular Value Decomposition (SVD) and retaining only the top-$k$ components, we effectively decouple true consensus from metric noise. Experiments demonstrate that SVD-MBR successfully regularizes decoding, yielding substantial gains across a range of generalized metrics. Furthermore, we reveal that this denoising is metric-dependent: neural metrics encode a robust low-rank consensus ideal for SVD, whereas surface-level metrics struggle to separate signal from metric noise.
CommentsAccepted to EMNLP 2026 Main
