From Static Templates to Dynamic Runtime Graphs: A Survey of Workflow Optimization for LLM Agents

• URL: http://arxiv.org/abs/2603.22386v1
• Date: Mon, 23 Mar 2026 17:42:59 GMT
• Title: From Static Templates to Dynamic Runtime Graphs: A Survey of Workflow Optimization for LLM Agents
• Abstract summary: Large language model (LLM)-based systems are becoming increasingly popular for solving tasks.<n>This survey reviews recent methods for designing and optimizing such computation.
• Score: 39.457720579458
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language model (LLM)-based systems are becoming increasingly popular for solving tasks by constructing executable workflows that interleave LLM calls, information retrieval, tool use, code execution, memory updates, and verification. This survey reviews recent methods for designing and optimizing such workflows, which we treat as agentic computation graphs (ACGs). We organize the literature based on when workflow structure is determined, where structure refers to which components or agents are present, how they depend on each other, and how information flows between them. This lens distinguishes static methods, which fix a reusable workflow scaffold before deployment, from dynamic methods, which select, generate, or revise the workflow for a particular run before or during execution. We further organize prior work along three dimensions: when structure is determined, what part of the workflow is optimized, and which evaluation signals guide optimization (e.g., task metrics, verifier signals, preferences, or trace-derived feedback). We also distinguish reusable workflow templates, run-specific realized graphs, and execution traces, separating reusable design choices from the structures actually deployed in a given run and from realized runtime behavior. Finally, we outline a structure-aware evaluation perspective that complements downstream task metrics with graph-level properties, execution cost, robustness, and structural variation across inputs. Our goal is to provide a clear vocabulary, a unified framework for positioning new methods, a more comparable view of existing body of literature, and a more reproducible evaluation standard for future work in workflow optimizations for LLM agents.

Related papers

• FlowScout: From Execution Feedback to Reliable Tool-Using Agent Workflows [6.767792769288717]
  FlowScout is an execution-guided framework for generating tool-integrated agentic from historical task-solving records.<n>It first mines a common tool coordination from historical records to construct an initial workflow.<n>It then refines the workflow topology through Monte Carlo tree search guided by execution feedback.
  arXiv  Detail & Related papers  (2026-08-10T08:41:59Z)
• Behavioral Controllability of Agentic Models for Information Extraction: From Fixed Workflows to Reflective Agents [4.2476244489062704]
  We study whether agentic components such as reflection and memory lead to observable and controllable improvements over fixed LLM agents.<n>We compare a fixed workflow baseline with reflective agent variants and specify an optimized agent condition.<n>Our evaluation emphasizes process-level behavior--including tool execution, retries, reflection, memory use, runtime failure recovery--while treating extraction coverage and field completeness as secondary outcome measures.
  arXiv  Detail & Related papers  (2026-07-17T07:51:09Z)
• FlowCompile: An Optimizing Compiler for Structured LLM Workflows [44.34248788469801]
  FlowCompile is a structured LLM compiler that performs compile-time design space exploration to identify a high-quality, reusable trade-off set.<n> Experiments show that FlowCompile consistently outperforms optimized workflow configurations and routing-based baselines, delivering up to 6.4x speedup.
  arXiv  Detail & Related papers  (2026-05-13T15:06:36Z)
• A Reproducible Optimisation Protocol for Calibrating Prompt-Based Large Language Model Workflows in Evidence Synthesis [1.624454100511275]
  This article presents a reproducible calibration workflow for prompt-based large language models (LLMs) in structured evidence-synthesis tasks.<n>The method separates the rules that define the scientific task from the mutable prompt harness that frames and applies them.<n>It optimises that harness against labelled or reference examples and an explicit task metric, then preserves the calibrated workflow as an inspectable artefact.
  arXiv  Detail & Related papers  (2026-05-07T20:55:07Z)
• Agentic Predictor: Performance Prediction for Agentic Workflows via Multi-View Encoding [56.565200973244146]
  Agentic Predictor is a lightweight predictor for efficient agentic workflow evaluation.<n>By learning to approximate task success rates, Agentic Predictor enables fast and accurate selection of optimal agentic workflow configurations.
  arXiv  Detail & Related papers  (2025-05-26T09:46:50Z)
• Understanding and Optimizing Agentic Workflows via Shapley value [49.508008396810624]
  We introduce ShapleyFlow, the first framework that employs cooperative game theory to analyze and optimize agentic configurations.<n>ShagleyFlow enables fine-grained attribution of each component's contribution and facilitates the identification of task-specific optimal configurations.
  arXiv  Detail & Related papers  (2025-02-01T18:07:34Z)
• Flow: Modularized Agentic Workflow Automation [53.073598156915615]
  Multi-agent frameworks powered by large language models (LLMs) have demonstrated great success in automated planning and task execution.<n>However, the effective adjustment of agentic during execution has not been well studied.<n>In this paper, we define an activity-on-vertex (AOV) graph, which allows continuous workflow refinement by agents.<n>Our proposed multi-agent framework achieves efficient concurrent execution of subtasks, effective goal achievement, and enhanced error tolerance.
  arXiv  Detail & Related papers  (2025-01-14T04:35:37Z)
• Opus: A Large Work Model for Complex Workflow Generation [0.0]
  Opus is a framework for generating and optimizing tasks tailored to complex Business Process Outsourcing (BPO) use cases.<n>Our approach generates executables from Intention, defined as the alignment of Client Input, Client Output and Process Directed Context.
  arXiv  Detail & Related papers  (2024-11-30T20:00:41Z)
• Benchmarking Agentic Workflow Generation [80.74757493266057]
  We introduce WorfBench, a unified workflow generation benchmark with multi-faceted scenarios and intricate graph workflow structures.<n>We also present WorfEval, a systemic evaluation protocol utilizing subsequence and subgraph matching algorithms.<n>We observe that the generated can enhance downstream tasks, enabling them to achieve superior performance with less time during inference.
  arXiv  Detail & Related papers  (2024-10-10T12:41:19Z)
• ComfyGen: Prompt-Adaptive Workflows for Text-to-Image Generation [87.39861573270173]
  We introduce the novel task of prompt-adaptive workflow generation, where the goal is to automatically tailor a workflow to each user prompt. We propose two LLM-based approaches to tackle this task: a tuning-based method that learns from user-preference data, and a training-free method that uses the LLM to select existing flows. Our work shows that prompt-dependent flow prediction offers a new pathway to improving text-to-image generation quality, complementing existing research directions in the field.
  arXiv  Detail & Related papers  (2024-10-02T16:43:24Z)

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.