Metraipolis › archive › dc_bd7d55d7edfb
LAGO: Language Agents with Reusable Growing Options — arXiv:2610.19843
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.19843
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T13:50:45.355Z
- SHA-256 of the archived text
b90413d5f236fe2c5981468eb5ac2c6e8f1e2d36ab27ca8df38b937b4525d78d- Stable URL of this page
https://metraipolis.com/documento/dc_bd7d55d7edfb
How to cite
LAGO: Language Agents with Reusable Growing Options — arXiv:2610.19843 — arXiv (en), https://arxiv.org/abs/2610.19843 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bd7d55d7edfb (Metraipolis archive dc_bd7d55d7edfb, sha256 b90413d5f236fe2c5981468eb5ac2c6e8f1e2d36ab27ca8df38b937b4525d78d)
BibTeX
@misc{dcbd7d55d7edfb,
title = {LAGO: Language Agents with Reusable Growing Options — arXiv:2610.19843},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.19843},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bd7d55d7edfb; sha256 b90413d5f236fe2c5981468eb5ac2c6e8f1e2d36ab27ca8df38b937b4525d78d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19843. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-horizon tasks expose a weakness of monolithic tool-use loops: one agent must decompose, execute, and verify a task within a single trajectory. We introduce LAGO, which treats specialized language agents as reusable temporally extended options. A meta-agent retrieves options from a persistent library or spawns new ones, dispatches sub-tasks, verifies results, and admits successful novel options. The library grows across tasks without updating model weights. On 500 SWE-bench Verified tasks, frozen Qwen3.5-35B-A3B in both roles resolves 58.6% under single-attempt evaluation with the official Docker evaluator. On a fixed 150-task subset, we re-run mini-SWE-agent, Terminus 2, and LAGO with the same policy, temperature, task IDs, and evaluator; their resolve rates are 49.3%, 57.3%, and 60.3%, respectively. Cross-model experiments provide evidence consistent with a role for decomposition quality. A 50-task component ablation finds four fewer resolved tasks without decomposition, two fewer with shuffled reused content, and six fewer without the library. These results support reusable agent-options as a practical abstraction for task streams, while leaving component-level causality task- and model-dependent. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Multiagent Systems (cs.MA). Pagina del lavoro: https://arxiv.org/abs/2610.19843 — 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 2026-10-15T13:48:35.460Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).