Zinuo You

Publications

Ideas, evidence,
and public artifacts.

Accepted and published work is labelled directly. Anonymous review work is described without naming the reviewing venue.

Google Scholar
2026
AAAI 2026 · published

How Wide and How Deep? Mitigating Over-squashing of GNNs via Channel Capacity Constrained Estimation

Zinuo You, Jin Zheng, John Cartlidge

An information-theoretic method for estimating GNN width and propagation depth under channel-capacity constraints.

Paper Code
2026
ICML 2026 Spotlight · accepted

Latent Laplace Diffusion for Irregular Multivariate Time Series

Zinuo You, Jin Zheng, John Cartlidge

Horizon-wide latent diffusion for irregular forecasting and imputation, using stable Laplace-domain modes and gap-aware conditioning.

Paper Code
2026
Manuscript · under-review

Refractive Clocks: Diffusion-to-Flow Schedule Transfer for Conditional Time Series Flow Matching

Zinuo You, Jin Zheng, John Cartlidge

A temporal-reparameterization view of transferring diffusion schedules into frozen flow-matching systems.

Code
2025
ICAAI 2025 · published

Improving Bayesian Optimization for Portfolio Management with Adaptive Scheduling

Zinuo You, John Cartlidge, Kevin Elliott, Meng Ge, David Gold

Budget-aware mean-variance scheduling and importance correction for sample-efficient black-box portfolio optimization.

Code
2025
LNCS / ICAART 2025 · published

Multi-Stock Movements Classification: A Decoupled Graph Diffusion Neural Network Approach

Zinuo You, Zhengyang Shi, Hao Bo, John Cartlidge, Liping Zhang, Yu Ge

Extended study of decoupled graph diffusion for multi-stock movement classification.

Code
2024
ICAART 2024 · published

DGDNN: Decoupled Graph Diffusion Neural Network for Stock Movement Prediction

Zinuo You, Zhengyang Shi, Hao Bo, John Cartlidge, Liping Zhang, Yu Ge

Decoupled propagation and transformation for representation learning over dynamic financial graphs.

Code
2024
ICASSP 2024 · published

Multi-Relational Graph Diffusion Neural Network with Parallel Retention for Stock Trends Classification

Zinuo You, Peng Zhang, Jin Zheng, John Cartlidge

Relation-aware graph diffusion with parallel retention for multi-stock trend classification.

Code