When does Subgroup Balance Matter? Predicting Performance Sensitivity from Latent Separability

Unequal representation of demographic groups in training data poses challenges to model generalisation across populations. Standard practice assumes that balancing subgroup representation always optimises performance. However, recent empirical results contradict this assumption: in some cases, imbalanced data distributions actually improve subgroup performance, and in others, subgroup performance is unaffected by the absence of an entire subgroup during training. We propose the latent separation hypothesis, which states that a fine-tuned model's dependence on subgroup representation is determined by the degree of separation between subgroups in the latent space of the pre-trained model. We formalise this hypothesis, provide theoretical analysis, validate it empirically across four vision and language datasets, and present preliminary evidence of a causal role through regularisation experiments. Finally, we show how quantitative analysis of latent subgroup separation can inform practical data collection and balancing decisions when fine-tuning foundation models.