Procedural Graphs for LLM Agents
Procedural Graphs: Self-Evolving Execution Structures for LLM Agents
Procedural Graphs improve LLM agent performance across multiple datasets and model types, offering a practical solution for developers building complex AI systems.
The paper introduces Procedural Graphs, a framework that organizes procedural knowledge into (procedure, relation, procedure) triplets for LLM agents. The framework localizes the agent's active node at each decision step, using a guidance model to translate subgraphs into situational guidance. The graph self-evolves through an LLM refiner that contrasts failed and successful trajectories to improve performance without manual engineering.
Procedural Graphs: Self-Evolving Execution Structures for LLM Agents
Large language models are increasingly deployed as agents that plan over long horizons and act through external tools. Most agents select actions through unconstrained generation over an accumulating history, leaving implicit the procedural knowledge of what to do, in what order, and under which conditions. As trajectories lengthen, agents can lose track of their objectives, invoke tools out of order, and repeat unproductive actions. We introduce the Procedural Graph: just as a knowledge graph organizes factual knowledge into (entity, relation, entity) triplets for what-is questions, a Procedural Graph organizes procedural knowledge into (procedure, relation, procedure) triplets for what-to-do questions. At each decision step, the framework localizes the agent's active node, and a guidance model translates the surrounding subgraph into step-level situational guidance that biases the solver's next action without dictating it. The graph is self-evolving: an LLM refiner contrasts failed trajectories with successful ones and edits the graph's topology and attributes, committing edits that preserve or improve held-out validation performance while retaining rejected ones to discourage repetition. Starting from a minimal skeleton, the loop builds graphs that match or surpass hand-designed ones. It can also repair a flawed expert prior. Across multiple datasets, task types, and LLMs, the Procedural Graph delivers consistent gains over memory-based baselines, and self-evolution further improves performance without manual engineering.
