IEEE

DisHelis: Optimizing Deployment of Disaggregated LLMs Inference Serving over Heterogeneous Environments via Hierarchical Max-Flow

Abstract: Disaggregated LLM inference service (DLIS), which decouples the compute-intensive prefill phase and the memory-intensive decode phase, enables more flexible and efficient resource usage.
IEEE

Solving the Problem of Vehicle Routing Problem With Time Window via Dual Adaptive Genetic Algorithm

Abstract: To address the Vehicle Routing Problem with Time Windows (VRPTW), this paper presents a novel Dual Adaptive Genetic Algorithm (DAGA). While numerous metaheuristic algorithms have been ...
Frontiers

Steady-state 1D two-phase flow differentiable modeling: learning from field data and inverse problem applications in oil wells

Accurately modeling steady-state two-phase flow is critical for the design and operation of systems in the oil and gas industry; however, traditional models often struggle to adapt to specific field ...