architectures
In the context of AI, architectures refer to the structural design of neural networks, defining how layers and components are organized and interact. Different architectures dictate the capabilities and performance of models for various tasks.
- $\boldsymbol{\lambda}$-Orthogonality Regularization for Compatible Representation Learning
- Knowledge Distillation Detection for Open-weights Models
- Learning Task-Agnostic Representations through Multi-Teacher Distillation
- Lost in Latent Space: An Empirical Study of Latent Diffusion Models for Physics Emulation
- The Quest for Universal Master Key Filters in DS-CNNs