activity
20242026
most citedSGFormer: Single-Layer Graph Transformers with Approximation-Free Linear Complexity

1 citations · 1 across the 6 of their papers we have counts for

collaborators

8 papers

math.OC2026

SUDA-Muon: Structural Design Principles and Boundaries for Fully Decentralized Muon

Hengrui Zhang, Boao Kong, Jiahe Geng +1

Fully decentralized Muon is difficult because its nonlinear matrix-sign operator does not commute with linear gossip averaging. This makes decentralized Muon a structural design pr…

math.OC2026

On the Convergence of Stochastic Gradient Descent with Perturbed Forward-Backward Passes

Boao Kong, Hengrui Zhang, Kun Yuan

We study stochastic gradient descent (SGD) for composite optimization problems with sequential operators subject to perturbations in both the forward and backward passes. Unlik…

cs.LG2025

A Diffusion-Based Method for Learning the Multi-Outcome Distribution of Medical Treatments

Yuchen Ma, Jonas Schweisthal, Hengrui Zhang +1

In medicine, treatments often influence multiple, interdependent outcomes, such as primary endpoints, complications, adverse events, or other secondary endpoints. Hence, to make op…

cs.CR2025

MUSE: Model-Agnostic Tabular Watermarking via Multi-Sample Selection

Liancheng Fang, Aiwei Liu, Henry Peng Zou +4

We introduce MUSE, a watermarking algorithm for tabular generative models. Previous approaches typically leverage DDIM invertibility to watermark tabular diffusion models, but tabu…

cs.LG2025

TabGen-ICL: Residual-Aware In-Context Example Selection for Tabular Data Generation

Liancheng Fang, Aiwei Liu, Hengrui Zhang +3

Large Language models (LLMs) have achieved encouraging results in tabular data generation. However, existing approaches require fine-tuning, which is computationally expensive. Thi…

cs.LG2024

Diffusion-nested Auto-Regressive Synthesis of Heterogeneous Tabular Data

Hengrui Zhang, Liancheng Fang, Qitian Wu +1

Autoregressive models are predominant in natural language generation, while their application in tabular data remains underexplored. We posit that this can be attributed to two fac…