learned representations
The high-level abstractions or features that a model has learned from data. These representations are crucial for tasks such as classification or generation, as they encapsulate the essential characteristics of the input data.
- $\boldsymbol{\lambda}$-Orthogonality Regularization for Compatible Representation Learning
- Automatic Visual Instrumental Variable Learning for Confounding-Resistant Domain Generalization
- Contrastive Representations for Temporal Reasoning
- Diffusion Feature Field for Text-based 3D Editing with Gaussian Splatting
- Diversity Is All You Need for Contrastive Learning: Spectral Bounds on Gradient Magnitudes
- Interpreting Emergent Features in Deep Learning-based Side-channel Analysis
- Joint‑Embedding vs Reconstruction: Provable Benefits of Latent Space Prediction for Self‑Supervised Learning
- Neural Collapse under Gradient Flow on Shallow ReLU Networks for Orthogonally Separable Data
- Obliviator Reveals the Cost of Nonlinear Guardedness in Concept Erasure
- Representational Difference Explanations
- The third pillar of causal analysis? A measurement perspective on causal representations
- Towards A Translative Model of Sperm Whale Vocalization
- Tracing the Representation Geometry of Language Models from Pretraining to Post-training