AI 中文总结
该研究提出DiffGBM模型，通过显式调整基于树的扩散与流模型的设计维度，在表格回归任务上实现优于基线的预测性能，同时提升采样效率与校准度。
AI 中文摘要
基于树的扩散模型可在无需神经密度估计器的情况下拟合灵活的表格回归条件预测分布，但它们继承了神经设置中的设计默认项——包括加噪路径、参数化、训练分布、特征、采样器等。我们表明这些默认项是关键约束：梯度提升集成实际上解决的是一个监督回归问题，其条件由这些默认项决定。我们提出DiffGBM，该模型在两个维度上将这些默认项显式化。其一，采用高斯路径流匹配训练器对$p(y \mid x)$建模，直接学习速度场并代数恢复分数，支持少步确定性常微分方程采样。其二，我们将分数侧方案——残差化、EDM风格预处理、对数-西格玛时间采样、噪声水平特征、损失加权及直方图分辨率——作为可联合调整的维度，置于共享的LightGBM框架上，而非固定的组合。此“分数灵活（score-flex）”空间将已发表的方案视为特例；在11个表格基准测试中，采用折-0调参、折1-5评估，且匹配40次试验预算与采样器的设置下，所选配置在所有数据集上均优于该基线（配对威尔科克森检验11/0，$p<10^{-3}$），且最佳的总体连续排名概率得分（CRPS）技能为0.725，优于基线的0.699。这两类配置各有优劣：score-flex通过随机采样器提升准确性，是最慢的配置；而流匹配是最快的采样器（比已发表基线快5.2倍），且是DiffGBM中校准最佳的配置。调优后的非扩散基线仍在个别数据集上胜出，且随机（$\varepsilon>0$）流采样器未在帕累托意义上优于确定性采样器。
英文摘要
Tree-based diffusion models fit flexible conditional predictive distributions for tabular regression without a neural density estimator, but they inherit their design defaults---noising path, parameterization, training distribution, features, sampler---from the neural setting. We show these defaults are the binding constraint: what a gradient-boosted ensemble actually solves is a supervised regression problem whose conditioning they determine. We present DiffGBM, which makes them explicit along two axes. First, a Gaussian-path flow-matching trainer for $p(y \mid x)$ that learns a velocity field directly and recovers the score algebraically, admitting few-step deterministic ODE sampling. Second, we expose the score-side recipe---residualization, EDM-style preconditioning, log-sigma time sampling, noise-level features, loss weighting, and histogram resolution---as jointly tunable axes over a shared LightGBM surface rather than one frozen bundle. This \emph{score-flex} space represents the published recipe as a special case; across eleven tabular benchmarks under fold-0 tuning, folds-1--5 evaluation, and a matched 40-trial budget and sampler, the selected configurations beat that baseline on \emph{every} dataset (paired Wilcoxon $11/0$, $p<10^{-3}$), with the best aggregate CRPS skill (0.725 vs.\ 0.699) of any row. The two rows are complementary: score-flex buys accuracy with a stochastic sampler and is the slowest row, while flow matching is the cheapest sampler ($5.2\times$ faster than the published baseline) and the best-calibrated DiffGBM row. Tuned non-diffusion baselines still win individual datasets, and stochastic ($\varepsilon>0$) flow samplers do not Pareto-dominate the deterministic corner.
Comments25 pages, 2 figures. Code: github.com/silaskoemen/diffgbm
