collaborators

7 papers

cs.LG2026

Stochastic and Non-local Closure Modeling for Nonlinear Dynamical Systems via Latent Score-based Generative Models

Xinghao Dong, Huchen Yang, Jin-Long Wu

We propose a latent score-based generative AI framework for learning stochastic, non-local closure models and constitutive laws in nonlinear dynamical systems of computational mech…

cs.MA2026

Active Learning for Communication Structure Optimization in LLM-Based Multi-Agent Systems

Huchen Yang, Xinghao Dong, Dan Negrut +1

Optimizing the communication structure of large language model based multi-agent systems (LLM-MAS) has been shown to improve downstream performance and reduce token usage. Existing…

cs.IT2026

Bayesian experimental design: grouped geometric pooled posterior via ensemble Kalman methods

Huchen Yang, Xinghao Dong, Jinlong Wu

Bayesian experimental design (BED) for complex physical systems is often limited by the nested inference required to estimate the expected information gain (EIG) or its gradients.…

cs.LG2026

Synergizing Transport-Based Generative Models and Latent Geometry for Stochastic Closure Modeling

Xinghao Dong, Huchen Yang, Jin-long Wu

Diffusion models recently developed for generative AI tasks can produce high-quality samples while still maintaining diversity among samples to promote mode coverage, providing a p…

cs.LG2026

Bayesian Experimental Design for Model Discrepancy Calibration: A Rivalry between Kullback--Leibler Divergence and Wasserstein Distance

Huchen Yang, Xinghao Dong, Jin-Long Wu

Designing experiments that systematically gather data from complex physical systems is central to accelerating scientific discovery. While Bayesian experimental design (BED) provid…

cs.LG2025

Bayesian Experimental Design for Model Discrepancy Calibration: An Auto-Differentiable Ensemble Kalman Inversion Approach

Huchen Yang, Xinghao Dong, Jin-Long Wu

Bayesian experimental design (BED) offers a principled framework for optimizing data acquisition by leveraging probabilistic inference. However, practical implementations of BED ar…