Latent Stochastic Interpolants

arXiv:2506.02276v2 Announce Type: replace Abstract: Stochastic Interpolants (SI) is a powerful framework for generative modeling, capable of flexibly transforming between two probability distributions. However, its use in jointly optimized latent variable models remains unexplored as it requires direct access to the samples from the two distributions. This work presents Latent Stochastic Interpolants (LSI) enabling joint learning in a latent space with end-to-end optimized encoder, decoder and latent SI models. We achieve this by developing a principled Evidence Lower Bound (ELBO) objective derived directly in continuous time. The joint optimization allows LSI to learn effective latent representations along with a generative process that transforms an arbitrary prior distribution into the encoder-defined aggregated posterior. LSI sidesteps the simple priors of the normal diffusion models and mitigates the computational demands of applying SI directly in high-dimensional observation spaces, while preserving the generative flexibility of the SI framework. We demonstrate the efficacy of LSI through comprehensive experiments on the standard large scale ImageNet generation benchmark.

Related

• Bridging the Gap between Learning and Inference for Diffusion-Based Molecule Generation
• Flow-based Generative Modeling of Potential Outcomes and Counterfactuals
• Interpolating Discrete Diffusion Models with Controllable Resampling
• Energy-Weighted Flow Matching: Unlocking Continuous Normalizing Flows for Efficient and Scalable Boltzmann Sampling
• Complex normalizing flows can be information Kahler-Ricci flows

Source: arXiv cs.LG | 2026-04-23