Search papers, labs, and topics across Lattice.
The authors investigate the limitations of text-to-image diffusion models in generating multiple objects by training models on the controlled mosaic dataset, which allows for disentangling data effects related to concept generalization and compositional generalization. They find that scene complexity, rather than concept imbalance, is the primary bottleneck, and that counting objects is particularly challenging in low-data regimes. Furthermore, compositional generalization degrades significantly as more concept combinations are withheld during training.
Diffusion models struggle with multi-object generation not because of imbalanced concept representation, but primarily due to scene complexity and a surprising difficulty in counting, especially when training data is limited.
Text-to-image diffusion models achieve impressive visual fidelity, yet they remain unreliable in multi-object generation. Despite extensive empirical evidence of these failures, the underlying causes remain unclear. We begin by asking how much of this limitation arises from the data itself. To disentangle data effects, we consider two regimes across different dataset sizes: (1) concept generalization, where each individual concept is observed during training under potentially imbalanced data distributions, and (2) compositional generalization, where specific combinations of concepts are systematically held out. To study these regimes, we introduce mosaic (Multi-Object Spatial relations, AttrIbution, Counting), a controlled framework for dataset generation. By training diffusion models on mosaic, we find that scene complexity plays a dominant role rather than concept imbalance, and that counting is uniquely difficult to learn in low-data regimes. Moreover, compositional generalization collapses as more concept combinations are held out during training. These findings highlight fundamental limitations of diffusion models and motivate stronger inductive biases and data design for robust multi-object compositional generation.
