latent variables
Latent variables are variables that are not directly observed but are inferred from observed data within a model. In AI, they are often employed in probabilistic models and can capture underlying structures or concepts essential for understanding complex data relationships.
- A Few Moments Please: Scalable Graphon Learning via Moment Matching
- A Latent Multilayer Graphical Model For Complex, Interdependent Systems
- Bifrost-1: Bridging Multimodal LLMs and Diffusion Models with Patch-level CLIP Latents
- Characterization and Learning of Causal Graphs from Hard Interventions
- Equivariant Eikonal Neural Networks: Grid-Free, Scalable Travel-Time Prediction on Homogeneous Spaces
- HoT-VI: Reparameterizable Variational Inference for Capturing Instance-Level High-Order Correlations
- Information-theoretic Generalization Analysis for VQ-VAEs: A Role of Latent Variables
- L$^2$M: Mutual Information Scaling Law for Long-Context Language Modeling
- Large Language Bayes
- Local Learning for Covariate Selection in Nonparametric Causal Effect Estimation with Latent Variables
- Online Time Series Forecasting with Theoretical Guarantees
- Reward-oriented Causal Representation Learning
- TARFVAE: Efficient One-Step Generative Time Series Forecasting via TARFLOW based VAE
- Unleashing the Potential of Multimodal LLMs for Zero-Shot Spatio-Temporal Video Grounding