Metraipolis › archive › dc_1268ee352bf1
GraphThink: Graph-Enhanced LLM Thinking for Long-Horizon Embodied Task Planning — arXiv:2608.07905
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.07905
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T09:13:04.428Z
- SHA-256 of the archived text
51f0126146826313cb3a6219d3081764b08b064464e43d100171b6506f88f450- Stable URL of this page
https://metraipolis.com/documento/dc_1268ee352bf1
How to cite
GraphThink: Graph-Enhanced LLM Thinking for Long-Horizon Embodied Task Planning — arXiv:2608.07905 — arXiv (en), https://arxiv.org/abs/2608.07905 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1268ee352bf1 (Metraipolis archive dc_1268ee352bf1, sha256 51f0126146826313cb3a6219d3081764b08b064464e43d100171b6506f88f450)
BibTeX
@misc{dc1268ee352bf1,
title = {GraphThink: Graph-Enhanced LLM Thinking for Long-Horizon Embodied Task Planning — arXiv:2608.07905},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.07905},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1268ee352bf1; sha256 51f0126146826313cb3a6219d3081764b08b064464e43d100171b6506f88f450},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.07905. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Embodied agents using LLM-based planners often struggle with physical hallucinations, poor generalization to long-horizon tasks, and lack of environmental awareness. We propose GraphThink, a novel framework that integrates a task graph to provide structured knowledge for robust planning and a scene graph to maintain environmental memory for event-driven replanning. Specifically, the task graph guides LLM thinking through contextual prompting and iterative refinement, effectively mitigating planning hallucinations. Furthermore, within the GRPO framework, the task graph offers delicate reward design to train the LLM planner, enhancing long-horizon planning capabilities and improving generalization. Finally, an event-driven replanning module, powered by the scene graph, enables closed-loop environment awareness and error correction. GraphThink achieves state-of-the-art performance on the ALFRED benchmark. In particular, our high-level planner surpasses leading API-based LLMs on both the validation set and held-out long-horizon tasks, underscoring its robust zero-shot and few-shot capabilities. Additional evaluations further demonstrate strong out-of-distribution generalization to novel tasks and environments. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-11 in cs.RO). Categorie arXiv: Artificial Intelligence (cs.AI) ; Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2608.07905 — 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 2026-10-19T09:11:58.820Z dalla pagina https://arxiv.org/catchup/cs.RO/2026-08-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).