autoregressive model
An autoregressive model is a type of statistical model that predicts future values based on past observations. In AI, autoregressive models are used in applications like time series forecasting and natural language processing, where the generation of sequences is crucial.
- AutoPartGen: Autoregressive 3D Part Generation and Discovery
- Autoregressive Motion Generation with Gaussian Mixture-Guided Latent Sampling
- Flatten Graphs as Sequences: Transformers are Scalable Graph Generators
- Overcoming Challenges of Long-Horizon Prediction in Driving World Models
- Simultaneous Modeling of Protein Conformation and Dynamics via Autoregression
- Vision as a Dialect: Unifying Visual Understanding and Generation via Text-Aligned Representations