Problem
Research questions and friction points this paper is trying to address.
class imbalance
single-cell foundation models
long-tail loss functions
rare cell populations
systematic failure
Rethinking Class Imbalance for Single-Cell Foundation Models: A Systematic Benchmark Across Architectures and Long-Tail Loss Functions
🤖 AI Summary
研究针对单细胞基础模型中罕见细胞类型分类不准确的问题，通过系统比较六种长尾损失函数在三种架构和三个数据集上的表现，提供了结合不平衡生物数据的实用指南。
📝 Abstract
Single-cell foundation models (scGPT, scBERT, Geneformer) achieve cell-type classification accuracy up to 97.5% in our experiments, yet this aggregate accuracy can mask systematic failure on rare, often disease-relevant cell populations that long-tail loss functions are widely assumed to address. We present a systematic benchmark of six long-tail loss functions (cross-entropy, weighted CE, class-balanced loss, focal loss, LDAM, logit-adjusted softmax) across three architectures and three datasets (Multiple Sclerosis, Zheng68K, human Pancreas), totaling 162 controlled training runs (3 backbones x 3 datasets x 6 losses x 3 seeds). The gap between overall accuracy, Macro-F1, and rare-class recall under plain cross-entropy is consistent across all nine (architecture, dataset) settings, driven by dataset structure rather than pretraining. Rare-class failure itself splits into two regimes with distinct embedding-geometry signatures, visible before any loss is chosen: some classes are recoverable by the right loss, while others retain linear separability yet are absorbed into unrelated classes' neighborhoods under every evaluated loss and architecture. Among the recoverable classes, the efficacy of reweighting is predicted by a class's absolute training-set size, rather than its share of the dataset or the dataset's overall imbalance ratio. Class-balanced loss and LDAM are the most consistent choices across all nine settings, while logit adjustment trades rare-class precision for recall rather than improving both. Our results give both a reusable benchmark and mechanism-grounded practical guidelines for combining foundation models with imbalanced biological data.
Innovation
Methods, ideas, or system contributions that make the work stand out.
long-tail loss functions
single-cell foundation models
rare-class recall
class-balanced loss
LDAM
🔎 Similar Papers
No similar papers found.
