Position: LLM Serving Needs Mathematical Optimization and Algorithmic Foundations, Not Just Heuristics

• URL: http://arxiv.org/abs/2605.01280v1
• Date: Sat, 02 May 2026 06:31:11 GMT
• Title: Position: LLM Serving Needs Mathematical Optimization and Algorithmic Foundations, Not Just Heuristics
• Abstract summary: LLM inference serving has outgrown generics and now demands mathematical optimization and algorithmic foundations.<n>These general-purpose policies ignore the distinctive structure of LLM inference--dynamically growing KV cache memory, prefill-decode phase asymmetry, unknown output lengths, and continuous constraints.<n>We contend that the field must develop algorithms with provable performance guarantees across diverse workloads, rather than models that may succeed in some scenarios but fail unpredictably in others.
• Score: 3.143753806123382
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: This position paper argues that LLM inference serving has outgrown generic heuristics and now demands mathematical optimization and algorithmic foundations. Despite rapid advances in serving systems such as vLLM and SGLang, their algorithmic cores remain largely unchanged from classical distributed computing: request routing uses join-shortest-queue or round-robin, scheduling defaults to FIFO, and KV cache eviction follows LRU. These general-purpose policies ignore the distinctive structure of LLM inference--dynamically growing KV cache memory, prefill-decode phase asymmetry, unknown output lengths, and continuous batching constraints. We contend that the field must develop mathematical models capturing these characteristics, enabling the design of algorithms with provable performance guarantees across diverse workloads, rather than heuristics that may succeed in some scenarios but fail unpredictably in others. Emerging work at the intersection of operations research and ML systems demonstrates that principled methods can match or exceed heuristic performance while providing theoretical guarantees. We call on the community to recognize algorithmic design for LLM serving as a research frontier.

Related papers

• FrontierOR: Benchmarking LLMs' Capacity for Efficient Algorithm Design in Large-Scale Optimization [61.43300970020897]
  Large language models (LLMs) are increasingly used for optimization modeling and solver-code generation.<n>Existing benchmarks are limited to small or simplified examples far below real-world scale and complexity.<n>We introduce FrontierOR, among the first benchmarks to evaluate LLM-based efficient algorithm design for realistic large-scale optimization problems.
  arXiv  Detail & Related papers  (2026-05-24T20:10:42Z)
• From Paper to Program: Accelerating Quantum Many-Body Algorithm Development via a Multi-Stage LLM-Assisted Workflow [4.866193347313755]
  Large language models (LLMs) can generate code rapidly but remain unreliable for scientific algorithms.<n>We introduce a multi-stage LLM-assisted workflow that separates theory extraction, formal specification, and code implementation.
  arXiv  Detail & Related papers  (2026-04-05T12:12:54Z)
• Cache Management for Mixture-of-Experts LLMs -- extended version [29.858964433575906]
  Large language models (LLMs) have demonstrated remarkable capabilities across a variety of tasks.<n>One of the main challenges towards the successful deployment of LLMs is memory management.<n>We introduce and study a new paging problem that models expert management optimization.
  arXiv  Detail & Related papers  (2025-09-02T15:19:06Z)
• LLM4CMO: Large Language Model-aided Algorithm Design for Constrained Multiobjective Optimization [54.35609820607923]
  Large language models (LLMs) offer new opportunities for assisting with algorithm design.<n>We propose LLM4CMO, a novel CMOEA based on a dual-population, two-stage framework.<n>LLMs can serve as efficient co-designers in the development of complex evolutionary optimization algorithms.
  arXiv  Detail & Related papers  (2025-08-16T02:00:57Z)
• Discrete Tokenization for Multimodal LLMs: A Comprehensive Survey [69.45421620616486]
  This work presents the first structured taxonomy and analysis of discrete tokenization methods designed for large language models (LLMs)<n>We categorize 8 representative VQ variants that span classical and modern paradigms and analyze their algorithmic principles, training dynamics, and integration challenges with LLM pipelines.<n>We identify key challenges including codebook collapse, unstable gradient estimation, and modality-specific encoding constraints.
  arXiv  Detail & Related papers  (2025-07-21T10:52:14Z)
• Optimizing LLM Inference: Fluid-Guided Online Scheduling with Memory Constraints [14.341123057506827]
  Large Language Models (LLMs) are indispensable in today's applications, but their inference procedure demands significant computational resources.<n>This paper formulates LLM inference optimization as a multi-stage online scheduling problem.<n>We develop a fluid dynamics approximation to provide a tractable benchmark that guides algorithm design.
  arXiv  Detail & Related papers  (2025-04-15T16:00:21Z)
• Throughput-Optimal Scheduling Algorithms for LLM Inference and AI Agents [6.318292471845427]
  We develop the queuing fundamentals for large language model (LLM) inference.<n>We prove that a large class of 'work-conserving' scheduling algorithms can achieve maximum throughput.
  arXiv  Detail & Related papers  (2025-04-10T00:12:12Z)
• Online Scheduling for LLM Inference with KV Cache Constraints [22.133592174540052]
  Large Language Model (LLM) inference is an intensive process requiring efficient scheduling to optimize latency and resource utilization.<n>We propose a novel and theoretically scheduling algorithm that minimizes inference latency while effectively managing the KV cache's memory.
  arXiv  Detail & Related papers  (2025-02-10T23:11:44Z)
• Designing Algorithms Empowered by Language Models: An Analytical Framework, Case Studies, and Insights [86.06371692309972]
  This work presents an analytical framework for the design and analysis of large language models (LLMs)-based algorithms.<n>Our proposed framework serves as an attempt to mitigate such headaches.
  arXiv  Detail & Related papers  (2024-07-20T07:39:07Z)
• A Dynamical Systems Approach for Convergence of the Bayesian EM Algorithm [59.99439951055238]
  We show how (discrete-time) Lyapunov stability theory can serve as a powerful tool to aid, or even lead, in the analysis (and potential design) of optimization algorithms that are not necessarily gradient-based. The particular ML problem that this paper focuses on is that of parameter estimation in an incomplete-data Bayesian framework via the popular optimization algorithm known as maximum a posteriori expectation-maximization (MAP-EM) We show that fast convergence (linear or quadratic) is achieved, which could have been difficult to unveil without our adopted S&C approach.
  arXiv  Detail & Related papers  (2020-06-23T01:34:18Z)
• Certified Reinforcement Learning with Logic Guidance [78.2286146954051]
  We propose a model-free RL algorithm that enables the use of Linear Temporal Logic (LTL) to formulate a goal for unknown continuous-state/action Markov Decision Processes (MDPs) The algorithm is guaranteed to synthesise a control policy whose traces satisfy the specification with maximal probability.
  arXiv  Detail & Related papers  (2019-02-02T20:09:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.