发表机构
George Mason University（乔治梅森大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出SwinDS-BWE，结合Swin-1D晶格生成器与三种决策科学判别器，将生成器参数降至17M、判别器降至1.36M，并在英法数据集上提升带宽扩展保真度。
AI 中文摘要
我们提出SwinDS-BWE，一种受决策科学启发的带宽扩展（BWE）模型，具有两项耦合贡献：（1）采用晶格风格跨流交互的Swin Transformer实现了局部感知建模，每个窗口的注意力成本随序列长度线性增长，并将先前AP-BWE生成器规模从33M缩减0.5倍至17M。（2）三个新颖的轻量级决策科学启发判别器增强了AP-BWE的性能：一个CVaR判别器对激活进行尾部池化以强调最差的高频（HF）片段，一个机会约束HF判别器通过可微屏障惩罚过量的HF能量，以及一个多准则效用判别器学习频谱准则上的凸风格权重。SwinDS-BWE超越了先前的AP-BWE，判别器规模缩小30倍（42.3M对比1.36M），并在英语和法语数据集上实现更高保真度。这项工作表明，受决策科学启发的批评者可以监督BWE以减少判别器规模。
英文摘要
We propose SwinDS-BWE, a decision-science-inspired bandwidth extension (BWE) model with two coupled contributions: (1) Swin Transformers with lattice-style cross-stream interaction yield locality-aware modeling with linear-in-sequence per-window attention cost and reduce the prior AP-BWE generator size 0.5x from 33M to 17M. (2) Three novel lightweight decision-science-inspired discriminators augment AP-BWE's performance: a CVaR discriminator tail-pools activations to emphasize worst high-frequency (HF) segments, a Chance-Constraint HF discriminator penalizes excessive HF energy via a differentiable barrier, and a Multi-Criteria Utility discriminator learns convex style weights over spectral criteria. SwinDS-BWE surpasses prior AP-BWE with a 30x smaller discriminator (42.3M vs. 1.36M) and higher fidelity on English and French datasets. This work shows that decision-science-inspired critics can supervise BWE to reduce discriminator size.
Comments5 pages, 1 figure, 4 tables; submitted to the 2027 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2027)
