Generative Modeling Comparison Hub
6 papers - avg viability 4.3
Recent advancements in generative modeling are focusing on enhancing robustness and efficiency across various applications. New frameworks, such as Conditional Unbalanced Optimal Transport, are addressing the challenges posed by outliers in conditional settings, which is crucial for tasks like image generation where data quality can vary significantly. Meanwhile, work on Fourier transformers is revolutionizing the discovery of crystalline materials by enabling the generation of complex structures while respecting physical constraints, thereby streamlining material science research. Additionally, the exploration of Wasserstein gradient flows is refining generative models to mitigate issues like mode collapse, enhancing their stability and performance. The shift towards Riemannian optimization in tensor networks is also noteworthy, as it improves the efficiency of generative modeling by leveraging manifold constraints. Collectively, these developments signal a maturation of the field, with a clear trajectory towards more reliable and application-ready generative models across diverse domains.
Top Papers
- Conditional Unbalanced Optimal Transport Maps: An Outlier-Robust Framework for Conditional Generative Modeling(7.0)
An outlier-robust conditional generative model that uses unbalanced optimal transport for improved distribution matching, offering a more reliable approach to conditional generation.
- Fourier Transformers for Latent Crystallographic Diffusion and Generative Modeling(5.0)
Innovative crystal generative model leveraging Fourier transforms and latent diffusion for material discovery.
- Gradient Flow Drifting: Generative Modeling via Wasserstein Gradient Flows of KDE-Approximated Divergences(4.0)
A new family of generative models leveraging Wasserstein gradient flows for improved performance in avoiding mode collapse.
- Efficient Generative Modeling with Unitary Matrix Product States Using Riemannian Optimization(4.0)
A novel Riemannian optimization approach for efficient generative modeling using unitary matrix product states.
- Generative Drifting is Secretly Score Matching: a Spectral and Variational Perspective(4.0)
A theoretical framework for generative modeling via drifting that enhances understanding and performance of image generation.
- On the Robustness of Langevin Dynamics to Score Function Error(2.0)
This paper analyzes the limitations of Langevin dynamics in score-based generative modeling.