Search papers, labs, and topics across Lattice.
CentraleSupélec, Gustave Roussy, Cancer Data Science Unit, Université Paris-Saclay, INSERM, IHU PRISM
3
0
4
Combining hard architectural sparsity with soft regularization can significantly enhance the interpretability of vision models without sacrificing performance.
Tile-level benchmarking can reliably shortlist top-performing models, potentially saving researchers significant time and resources in digital pathology.
Medical vision-language models perform better when the modality gap is tuned to an intermediate level, challenging the assumption that minimizing it is always optimal.
