arXiv subjects

Lele Li

Publications and source records attributed to Lele Li.

4 recordsLinked to original sources

arXiv2026

ASAP: A Disaggregated and Asynchronous Inference System for MoE Prefill

Mixture-of-Experts (MoE) models have become the de facto standard for scaling large language models. To maintain computational efficiency, modern MoE serving systems typically employ a hybrid parallelism strategy, combining Data Parallelism (DP) for attention stages with Expert Parallelism (EP) for MoE stages. However, this design necessitates frequent global synchronization barriers between attention DP groups and experts. In online serving, significant variance in request arrival rates and sequence lengths inherently leads to DP imbalance, causing severe synchronization stalls that degrade Time-to-First-Token (TTFT) and system throughput. We present ASAP, an asynchronous inference system specifically designed to accelerate the prefill phase of MoE models. ASAP disaggregates the attention and MoE stages and implements a fully asynchronous execution pipeline. This is achieved through a suite of specialized asynchronous communication primitives and four coordinated optimizations across request scheduling and model execution, which collectively dismantle global synchronization barriers. We implement and evaluate ASAP on CloudMatrix384 super-nodes, demonstrating that it improves SLO-compliant prefill throughput by 90% compared to state-of-the-art synchronous serving solutions.

cs.DC↗

arXiv2025

Generalized spectral characterization of signed bipartite graphs

Let $\Sigma$ be an $n$-vertex controllable or almost controllable signed bipartite graph, and let $\Delta_\Sigma$ denote the discriminant of its characteristic polynomial $\chi(\Sigma; x)$. We prove that if (\rmnum{1}) the integer $2^{ -\lfloor n/2 \rfloor }\sqrt{\Delta _{\Sigma}}$ is squarefree, and (\rmnum{2}) the constant term (even $n$) or linear coefficient (odd $n$) of $\chi(\Sigma; x)$ is $\pm 1$, then $\Sigma$ is determined by its generalized spectrum. This result extends a recent theorem of Ji, Wang, and Zhang [Electron. J. Combin. 32 (2025), \#P2.18], which established a similar criterion for signed trees with irreducible characteristic polynomials.

math.CO↗

arXiv2024

Towards Automatic Linearization via SMT Solving

Mathematical optimization is ubiquitous in modern applications. However, in practice, we often need to use nonlinear optimization models, for which the existing optimization tools such as Cplex or Gurobi may not be directly applicable and an (error-prone) manual transformation often has to be done. Thus, to address this issue, in this paper we investigate the problem of automatically verifying and synthesizing reductions, the solution of which may allow an automatic linearization of nonlinear models. We show that the synthesis of reductions can be formulated as an $\exists^* \forall^*$ synthesis problem, which can be solved by an SMT solver via the counter-example guided inductive synthesis approach (CEGIS).

cs.LO↗

arXiv2023

Scheduling Distributed Flexible Assembly Lines using Safe Reinforcement Learning with Soft Shielding

Highly automated assembly lines enable significant productivity gains in the manufacturing industry, particularly in mass production condition. Nonetheless, challenges persist in job scheduling for make-to-job and mass customization, necessitating further investigation to improve efficiency, reduce tardiness, promote safety and reliability. In this contribution, an advantage actor-critic based reinforcement learning method is proposed to address scheduling problems of distributed flexible assembly lines in a real-time manner. To enhance the performance, a more condensed environment representation approach is proposed, which is designed to work with the masks made by priority dispatching rules to generate fixed and advantageous action space. Moreover, a Monte-Carlo tree search based soft shielding component is developed to help address long-sequence dependent unsafe behaviors and monitor the risk of overdue scheduling. Finally, the proposed algorithm and its soft shielding component are validated in performance evaluation.

cs.NE↗