LLM Agent Workflow Optimization

LLM: Large Language Model

This field studies how to organize language models, tools, memory, and decision steps so agents can complete multi-step tasks effectively and efficiently. Approaches include fixed workflows and DAGs, query-level model routing, reinforcement learning and evolutionary search, and trace-guided prompt or harness refinement; they are tested in settings such as coding, web use, retrieval, and scientific research. Systems work also optimizes execution through caching, scheduling, parallelism, and compiling repeated behavior into deterministic workflows. A central tension is whether adaptive workflows improve on reusable fixed designs enough to justify their added cost and complexity, while maintaining reliability and generalizing beyond the tasks used for optimization.

Latest papers

262 papers59 papers in the last four weeks, up 69% on the four weeks before. 0.3% of all new papers.

Build, Judge, Optimize: A Blueprint for Continuous Improvement of Multi-Agent Consumer Assistants
Mar 3, 2026Alejandro Breen Herrera, Aayush Sheth, Steven G. Xu +8Multi-Agent LLM Systems LLM Agent Workflow Optimization
Metis · DoorDash
DenoiseFlow: Uncertainty-Aware Denoising for Reliable LLM Agentic Workflows
Feb 28, 2026Yandong Yan, Junwei Peng, Shijie Li +7LLM Agent Reliability LLM Agent Workflow Optimization
School of Computer Science, Peking University · School of Electronics Engineering and Computer Science, Peking University · China Southern Power Grid +6
WirelessAgent++: Automated Agentic Workflow Design and Benchmarking for Wireless Networks
Feb 28, 2026Jingwen Tong, Zijian Li, Fang Liu +2Wireless Local Area Networks Agentic Workflows
College of Electronics and Information Engineering, Shenzhen University, Shenzhen, China · Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology, Hong Kong SAR
AI Runtime Infrastructure
Feb 28, 2026Christopher CruzRuntime Enforcement for AI Agents LLM Agent Workflow Optimization
Toward Expert Investment Teams:A Multi-Agent LLM System with Fine-Grained Trading Tasks
Feb 26, 2026Kunihiro Miyazaki, Takanobu Kawahara, Stephen Roberts +1Multi-Agent LLM Systems Algorithmic Trading
Japan Digital Design, Inc. Tokyo, Japan · Department of Engineering Science University of Oxford United Kingdom
MiroFlow: Towards High-Performance and Robust Open-Source Agent Framework for General Deep Research Tasks
Feb 26, 2026Shiqian Su, Sen Xing, Xuan Dong +13Autonomous Research Agents LLM Agent Orchestration
Tsinghua University · MiroMind AI · National University of Singapore +1
Human-Guided Agentic AI for Multimodal Clinical Prediction: Lessons from the AgentDS Healthcare Benchmark
Feb 23, 2026Lalitha Pranathi Pulavarthy, Raajitha Muthyala, Aravind V Kuruvikkattil +3Human-AI Collaboration Clinical Prediction
Dept. Biomedical Eng. & Informatics, Indiana University Indianapolis, Indianapolis, USA
Robust and Efficient Tool Orchestration via Layered Execution Structures with Reflective Correction
Feb 21, 2026Tao Zhe, Haoyu Wang, Bo Luo +6LLM Tool Use LLM Agent Orchestration
Department of Electrical Engineering and Computer Science, University of Kansas, Lawrence, KS, USA · NEC Laboratories America, Princeton, NJ, USA · Institute for Infocomm Research, A*STAR, Singapore +2
Alignment in Time: Peak-Aware Orchestration for Long-Horizon Agentic Systems
Feb 20, 2026Hanjing Shi, Dominic DiFranzoLong-Horizon LLM Agents LLM Agent Workflow Optimization
Lehigh University, Bethlehem, Pennsylvania, USA
AgentConductor: Topology Evolution for Multi-Agent Competition-Level Code Generation
Feb 19, 2026Siyu Wang, Ruotian Lu, Zhihao Yang +7LLM Agent Workflow Optimization Multi-Agent Coding
Shanghai Jiao Tong University · Meituan
TabAgent: Distilling LLM Agent Components into Cheap Classifiers
Feb 18, 2026Ido Levy, Segev Shlomov, Avi Yaeli +5LLM Agent Workflow Optimization AI Agent Architecture
IBM Research, Haifa, Israel
Optimization Instability in Autonomous Agentic Workflows for Clinical Symptom Detection
Feb 17, 2026Cameron Cagan, Pedram Fard, Jiazi Tian +3LLM Agent Workflow Optimization AI Agent Reliability
Department of Medicine, Massachusetts General Hospital, Boston, MA, USA · Department of Neurology, Massachusetts General Hospital, Boston, MA, USA
Beyond Static Pipelines: Learning Dynamic Workflows for Text-to-SQL
Feb 17, 2026Yihan Wang, Peiyu Liu, Runyu Chen +1Text-to-SQL LLM Agent Workflow Optimization
Renmin University of China · University of International Business and Economics
Multi-Modal Multi-Agent Reinforcement Learning for Radiology Report Generation
Feb 17, 2026Kaito Baba, Risa Kishikawa, Satoshi KoderaRadiology Report Generation Multimodal Agents
Department of Cardiovascular Medicine The University of Tokyo Hospital, Tokyo, Japan
Arbor: A Framework for Reliable Navigation of Critical Conversation Flows
Feb 16, 2026Luís Silva, Diogo Gonçalves, Catarina Farinha +2LLM Agent Workflow Optimization Efficient Language Model Inference
Sword Health
ThunderAgent: A Simple, Fast and Program-Aware Agentic Inference System
Feb 14, 2026Hao Kang, Ziyang Li, Weili Xu +7LLM Agent Orchestration LLM Inference Scheduling
University of Illinois Urbana-Champaign · 5Together AI · 1Georgia Institute of Technology
Learning to Configure Agentic AI Systems
Feb 12, 2026Aditya Taparia, Som Sagar, Ransalu SenanayakeLLM Agent Workflow Optimization Agentic RL
School of Computing and Augmented Intelligence Arizona State University Tempe, AZ, USA
ReCreate: Reasoning and Creating Domain Agents Driven by Experience
Jan 16, 2026Zhezheng Hao, Hong Wang, Jian Luo +6LLM Agents LLM Agent Self-Improvement
Zhejiang University · State Key Laboratory of Blockchain and Data Security, Zhejiang University · Tencent +3
Nalar: Workflow-Aware Management of Agentic Applications
Jan 8, 2026Saurabh Agarwal, Marco Laju, Donghyun Son +4LLM Inference Efficiency LLM Agent Orchestration
UT-Austin · Cisco-Research
Rethinking Query Optimization for Multi-Agent Systems [Vision]
Dec 10, 2025Zoi Kaoudi, Ioana GiurgiuMulti-Agent System Optimization Multi-Agent Orchestration
IT University of Copenhagen · IBM Research Europe
Evolving Excellence: Automated Optimization of LLM-based Agents
Dec 9, 2025Paul Brookes, Vardan Voskanyan, Rafail Giavrimis +18Large Language Model-Guided Optimization LLM Agent Workflow Optimization
TurinTech AI, London, UK · University of Surrey, Guildford, United Kingdom · University of Leeds, Leeds, UK +4
AutoMind: Adaptive Knowledgeable Agent for Automated Data Science
Jun 12, 2025Yujie Luo, Yixin Ou, Jingsheng Zheng +9Data Analysis Agents LLM Agents
Zhejiang University · Zhejiang University - Ant Group Joint Laboratory of Knowledge Graph · Ant Group

Mar 3, 2026J/K move · Enter open · S save

Build, Judge, Optimize: A Blueprint for Continuous Improvement of Multi-Agent Consumer Assistants

Alejandro Breen Herrera, Aayush Sheth, Steven G. Xu, Zhucheng Zhan+7

Metis · DoorDash

Multi-Agent LLM Systems LLM Agent Workflow Optimization Agentic Commerce Prompt Optimization Multi-Turn Dialogue Evaluation

Open paper AskPDF arXiv

Conversational shopping assistants (CSAs) represent a compelling application of agentic AI, but moving from prototype to production reveals two underexplored challenges: how to evaluate multi-turn interactions and how to optimize tightly coupled multi-agent systems. Grocery shopping further amplifies these difficulties, as user requests are often underspecified, highly preference-sensitive, and constrained by factors such as budget and inventory. In this paper, we present a practical blueprint for evaluating and optimizing conversational shopping assistants, illustrated through a production-scale AI grocery assistant. We introduce a multi-faceted evaluation rubric that decomposes end-to-end shopping quality into structured dimensions and develop a calibrated LLM-as-judge pipeline aligned with human annotations. Building on this evaluation foundation, we investigate two complementary prompt-optimization strategies based on a SOTA prompt-optimizer called GEPA (Shao et al., 2025): (1) Sub-agent GEPA, which optimizes individual agent nodes against localized rubrics, and (2) MAMuT (Multi-Agent Multi-Turn) GEPA (Herrera et al., 2026), a novel system-level approach that jointly optimizes prompts across agents using multi-turn simulation and trajectory-level scoring. We release rubric templates and evaluation design guidance to support practitioners building production CSAs.