Benjamin Wild

Geometry-preserving modifications to encoder architectures can dramatically enhance multimodal alignment, outperforming traditional expressive objectives.

Explicitly modeling intra- and cross-modal interactions in brain MRI can lead to substantial performance improvements, with MICViT outperforming traditional methods by leveraging multimodal inputs.

Multimodal models like Symile can appear to outperform CLIP even when a single unreliable modality silently corrupts the product terms, but a simple gating mechanism can fix this.