arXiv:2604.05906v1 Announce Type: cross
Abstract: Numerous studies on text-to-image (T2I) generative models have utilized cross-attention maps to boost application performance and interpret model behavior. However, the distinct characteristics of attention maps from different attention heads remain relatively underexplored. In this study, we show that selectively aggregating cross-attention maps from heads most relevant to a target concept can improve visual interpretability. Compared to the diffusion-based segmentation method DAAM, our approach achieves higher mean IoU scores. We also find that the most relevant heads capture concept-specific features more accurately than the least relevant ones, and that selective aggregation helps diagnose prompt misinterpretations. These findings suggest that attention head selection offers a promising direction for improving the interpretability and controllability of T2I generation.
Bioethical considerations in deploying mobile mental health apps in LMIC settings: insights from the MITHRA pilot study in rural India
IntroductionIn India, untreated depression among women contributes significantly to morbidity and mortality, underscoring an urgent need for accessible and ethically grounded mental health interventions. Mobile



