Metraipolis › archive › dc_fd84f895069f

Procedural Graphs: Self-Evolving Execution Structures for LLM Agents — arXiv:2609.09153

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.09153
License (as archived)
CC0-1.0
Archived
2026-10-16T12:58:43.914Z
SHA-256 of the archived text
0455a4fae564155220cb189940cde5b74afc50596ed7d5a57051f656f155ae7e
Stable URL of this page
https://metraipolis.com/documento/dc_fd84f895069f

How to cite

Procedural Graphs: Self-Evolving Execution Structures for LLM Agents — arXiv:2609.09153 — arXiv (en), https://arxiv.org/abs/2609.09153 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fd84f895069f (Metraipolis archive dc_fd84f895069f, sha256 0455a4fae564155220cb189940cde5b74afc50596ed7d5a57051f656f155ae7e)
BibTeX
@misc{dcfd84f895069f,
  title = {Procedural Graphs: Self-Evolving Execution Structures for LLM Agents — arXiv:2609.09153},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.09153},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fd84f895069f; sha256 0455a4fae564155220cb189940cde5b74afc50596ed7d5a57051f656f155ae7e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.09153. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CL).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2609.09153 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T12:30:25.394Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).