Joshua M. Susskind

To Infinity and Beyond: Tool-Use Unlocks Length Generalization in State Space Models

Adapting Self-Supervised Representations as a Latent Space for Efficient Generation

Rethinking JEPA: Compute‑Efficient Video Self-Supervised Learning with Frozen Teachers

SimpleFold: Folding Proteins is Simpler than You Think

SimpleDesign - A Joint Model for Protein Sequence and Structure Codesign

Inferring Attribute Subspaces from Visual Contexts

End-to-End Video Generative Modeling with Scalable Normalizing Flows

STARFlow: Scaling Latent Normalizing Flows for High-resolution Image Synthesis

Normalizing Flows are Capable Generative Models

TADA: Improved Diffusion Sampling with Training-free Augmented DynAmics

Flexible Language Modeling in Continuous Space with Transformer-based Autoregressive Flows

Proxy-FDA: Proxy-based Feature Distribution Alignment for Fine-tuning Vision Foundation Models without Forgetting

Target Concrete Score Matching: A Holistic Framework for Discrete Diffusion

Denoising Autoregressive Transformers for Scalable Text-to-Image Generation

INRFlow: Flow Matching for INRs in Ambient Space

Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models

Proxy-FDA: Proxy-based Feature Distribution Alignment for Fine-tuning Vision Foundation Models without Forgetting

Mechanisms of Projective Composition of Diffusion Models

Coordinate In and Value Out: Training Flow Transformers in Ambient Space

The Geometry of Attention: Ricci Curvature and Transformers Training and Robustness

Distilled Diffusion Language Models