发表机构
University of California, San Diego（加利福尼亚大学圣迭戈分校）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对BiomeGPT式微生物组Transformer的注意力权重局限，提出符号化集成梯度归因方法，结合二阶集成海森矩阵，可区分致病与保护性微生物信号，揭示群落相互作用规则，提升模型可解释性
AI 中文摘要
在特征标记化Transformer（arXiv:2106.11959）如BiomeGPT（doi: https://doi.org/10.64898/2026.01.05.697599）中，每个输入标记通过固定标识与样本特异性测量值融合构建：固定物种与可变丰度，即T = S + A。为解释此类模型的下游分类，现有研究通过检查特殊<[BOS_never_used_51bce0c785ca2f68081bfa7d91973934]>标记的注意力权重（arXiv:2106.11959、arXiv:1810.04805、BiomeGPT）对样本标记按重要性排序。这些权重存在两个关键局限：它们是非负的，因此无法区分疾病支持证据与健康支持证据（arXiv:2201.12114）；且它们在标记融合后作用，掩盖了输入源S和A各自对输出的影响。为解决此问题，我们采用集成梯度（arXiv:1703.01365）——一种符号化、感知融合的归因方法，并为BiomeGPT等特征标记化模型提出源派生基线T' = S + A₀，其将物种标识保留为固定生物坐标，同时隔离丰度变化的影响。将该方法应用于疾病-健康决策边际，它产生的极性可明确区分致病微生物信号与保护性微生物信号。我们表明，这种基于梯度的方法可揭示物种-丰度方向关系和敏感性诊断，而这些完全被非负的<[BOS_never_used_51bce0c785ca2f68081bfa7d91973934]>注意力权重所掩盖。我们进一步推荐使用二阶集成海森矩阵（arXiv:2002.04138）以揭示微生物组群落相互作用规则：一个成员的扰动如何改变模型对另一个成员的敏感性，以及在给定丰度水平下哪些其他物种将模糊案例推向疾病或健康。这为BiomeGPT提供了一种原则性的可解释性方法，可推广到其他平滑且可微的特征标记化Transformer。代码可在此URL获取
英文摘要
In a feature-tokenized transformer (arXiv:2106.11959) such as BiomeGPT (doi:10.64898/2026.01.05.697599), each input token is built by fusing a fixed identity with a sample-specific measurement: a fixed species and a variable abundance, T = S + A. To interpret downstream classification in such models, prior work inspects the attention weights of the special [CLS] token (arXiv:2106.11959, arXiv:1810.04805, BiomeGPT) to rank sample tokens by importance. These weights have two critical limitations: they are nonnegative, so they cannot separate disease-supporting from health-supporting evidence (arXiv:2201.12114), and they act after token fusion, obscuring how the input sources S and A each affect the output. To address this we use Integrated Gradients (arXiv:1703.01365), a signed, fusion-aware attribution method, and propose a source-derived baseline T' = S + A_0 for feature-tokenized models such as BiomeGPT, which preserves species identity as a fixed biological coordinate while isolating the effect of abundance variation. Applied to a disease-versus-health decision margin, it yields polarity that explicitly separates pathogenic from protective microbial signals. We show that this gradient-based approach uncovers species-abundance directional relationships and sensitivity diagnostics entirely obscured by unsigned [CLS] attention weights. We further recommend second-order Integrated Hessians (arXiv:2002.04138) to expose microbiome community interaction rules: how a perturbation in one member alters the model's sensitivity to another, and which other species drive ambiguous cases toward disease or health at a given abundance level. This provides a principled approach to explainability in BiomeGPT that generalizes to other smooth and differentiable feature-tokenized transformers. Code is available at https://github.com/nohren/token-source-attribution
Comments15 pages, 6 figures. Code: https://github.com/nohren/token-source-attribution
