Metraipolis › archive › dc_693183dd78a7
When Do Causal World Models Help Modular LLM Agents — arXiv:2610.00012
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.00012
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T12:01:08.551Z
- SHA-256 of the archived text
13c5401ffc173e61b2a25c04d4202ad8ded8aac664ee1af7523fe1590614a728- Stable URL of this page
https://metraipolis.com/documento/dc_693183dd78a7
How to cite
When Do Causal World Models Help Modular LLM Agents — arXiv:2610.00012 — arXiv (en), https://arxiv.org/abs/2610.00012 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_693183dd78a7 (Metraipolis archive dc_693183dd78a7, sha256 13c5401ffc173e61b2a25c04d4202ad8ded8aac664ee1af7523fe1590614a728)
BibTeX
@misc{dc693183dd78a7,
title = {When Do Causal World Models Help Modular LLM Agents — arXiv:2610.00012},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.00012},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_693183dd78a7; sha256 13c5401ffc173e61b2a25c04d4202ad8ded8aac664ee1af7523fe1590614a728},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00012. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
LLM agents increasingly act through modular systems, such as order, payment, inventory, and shipment services, where actions in one module change which transitions are valid in another. Standard world models usually fit observational traces, but this is not the quantity needed for intervention-time planning: a trace may show that payment precedes shipment without identifying whether payment authorizes shipment, inventory mediates the effect, or a hidden trigger explains both. We study this gap through FedCausalCompose, a causal world-model framework for modular LLM agents in which local actions provide intervention-response evidence for cross-module interfaces. We first show that observational world models incur an irreducible interventional error under unblocked back-door paths, that interface recovery improves with intervention-response coverage, and that an oracle causal composition can beat the non-causal lower bound when coverage and local mechanism errors are controlled. We then test the resulting prediction in diagnostic agent settings. Causal interfaces help most in structured tool environments, where API signatures expose preconditions and downstream effects. In contrast, dialogue and narrative environments often ignore raw edge lists unless a short attention anchor makes the causal information decision-relevant. These results identify a concrete condition for causal world models in LLM agents: causal structure helps when cross-module interfaces are both statistically identifiable and presented in a form the agent can use at action time. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.00012 — 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 2026-10-06T10:21:27.832Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).