9 papers
AdaHVLA: Adaptive Harnesses for Long-Horizon Vision-Language-Action Execution
Junyi Tang, Jie Peng, Zezhen Ding +2
Vision-language-action (VLA) models offer strong local control and instruction following but often struggle with long-horizon tasks requiring persistent memory and planning. Task h…
Learning to Price and Stock Under Contextual and Censored Demand
Zean Han, Zezhen Ding, Jiheng Zhang
To make optimal joint pricing and inventory control decisions is a critical challenge for modern retailers. In practice, retailers face changing market conditions where demands are…
Nonparametric Contextual Pricing and Inventory Learning under Censored Demand
Zean Han, Jing Liang, Ruihan Lin +2
In online retailing, when a product sells out, a retailer often sees only the units sold, not how many customers would have bought it had inventory been available. However, the inv…
Direction-Aware Offline-to-Online Learning in Linear Contextual Bandits
Zean Han, Ruihan Lin, Zezhen Ding +1
Many bandit systems are deployed with offline historical data, such as past logs from earlier policies. Using these data can reduce early online exploration when they remain inform…
A Kinetic Energy Perspective of Flow Matching
Ziyun Li, Huancheng Hu, Soon Hoe Lim +6
Flow-based generative models can be viewed through a physics lens: sampling transports a particle from noise to data by integrating a learned velocity field, and each sample corres…
Large-Scale LLM Inference with Heterogeneous Workloads: Prefill-Decode Contention and Asymptotically Optimal Control
Ruihan Lin, Zezhen Ding, Zean Han +1
Large Language Models (LLMs) are rapidly becoming critical infrastructure for enterprise applications, driving unprecedented demand for GPU-based inference services. A key operatio…
OR-R1: Automating Modeling and Solving of Operations Research Optimization Problem via Test-Time Reinforcement Learning
Zezhen Ding, Zhen Tan, Jiheng Zhang +1
Optimization modeling and solving are fundamental to the application of Operations Research (OR) in real-world decision making, yet the process of translating natural language prob…
Make Optimization Once and for All with Fine-grained Guidance
Mingjia Shi, Ruihan Lin, Xuxi Chen +8
Learning to Optimize (L2O) enhances optimization efficiency with integrated neural networks. L2O paradigms achieve great outcomes, e.g., refitting optimizer, generating unseen solu…
FlowTS: Time Series Generation via Rectified Flow
Yang Hu, Xiao Wang, Zezhen Ding +7
Diffusion-based models have significant achievements in time series generation but suffer from inefficient computation: solving high-dimensional ODEs/SDEs via iterative numerical s…