activity
20122026
most citedScore-Based Generative Modeling through Stochastic Differential Equations

1.3k citations · 4.9k across the 223 of their papers we have counts for

collaborators
Showing 2026 · cs.LGShow all

10 papers · 2 filters

cs.LG2026

Constrained Decoding for Diffusion Language Models via Efficient Inference over Finite Automata

Meihua Dang, Stefano Ermon

Constrained decoding is essential for serving LLMs, ensuring that generated outputs follow specific structures such as JSON schema-formatted function calls. Existing systems are de…

cs.LG2026

One-Step Generative Modeling via Wasserstein Gradient Flows

Jiaqi Han, Puheng Li, Qiushan Guo +3

Diffusion models and flow-based methods have shown impressive generative capability, especially for images, but their sampling is expensive because it requires many iterative updat…

cs.LG2026

Generative Modeling with Flux Matching

Peter Pao-Huang, Xiaojie Qiu, Stefano Ermon

We introduce Flux Matching, a new paradigm for generative modeling that generalizes existing score-based models to a broader family of vector fields that need not be conservative.…

cs.LG2026

Beyond Pairwise Preferences: Listwise Reward-Aware Alignment for Diffusion Models

Austin Wang, Jiaqi Han, Stefano Ermon +1

Preference optimization has emerged as an efficient alternative to online reinforcement learning from human feedback (RLHF) for aligning text-to-image diffusion models. However, ex…

cs.LG2026

Align Your Structures: Generating Trajectories with Structure Pretraining for Molecular Dynamics

Aniketh Iyengar, Jiaqi Han, Pengwei Sun +3

Generating molecular dynamics (MD) trajectories using deep generative models has attracted increasing attention, yet remains inherently challenging due to the limited availability…

cs.LG2026

ABC: Any-Subset Autoregression via Non-Markovian Diffusion Bridges in Continuous Time and Space

Gabe Guo, Thanawat Sornwanee, Lutong Hao +3

Generating continuous-time, continuous-space stochastic processes (e.g., videos, weather forecasts) conditioned on partial observations (e.g., first and last frames) is a fundament…