Metraipolis › archive › dc_2a6e9444aaee
Neuro-Symbolic Computer Use: Learning Reusable Policies for Reliable and Efficient Execution — arXiv:2609.36927
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.36927
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:23:16.120Z
- SHA-256 of the archived text
df257b87aa2636ca600866cc5af2785d80df680ae9b2a2e6c7efe5591e717c7b- Stable URL of this page
https://metraipolis.com/documento/dc_2a6e9444aaee
How to cite
Neuro-Symbolic Computer Use: Learning Reusable Policies for Reliable and Efficient Execution — arXiv:2609.36927 — arXiv (en), https://arxiv.org/abs/2609.36927 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2a6e9444aaee (Metraipolis archive dc_2a6e9444aaee, sha256 df257b87aa2636ca600866cc5af2785d80df680ae9b2a2e6c7efe5591e717c7b)
BibTeX
@misc{dc2a6e9444aaee,
title = {Neuro-Symbolic Computer Use: Learning Reusable Policies for Reliable and Efficient Execution — arXiv:2609.36927},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.36927},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2a6e9444aaee; sha256 df257b87aa2636ca600866cc5af2785d80df680ae9b2a2e6c7efe5591e717c7b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.36927. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Many computer tasks recur: the same workflow runs many times, with new inputs and from different starting states. Current computer-use agents re-plan every step of every run, which makes them costly and unreliable on such tasks. We introduce neuro-symbolic computer use, in which a recurring workflow is executed by a learned policy rather than re-derived by an agent on each run. The policy fixes the decisions that are stable across runs (ordering, variables, loops, and branches) in executable code, and delegates observation-dependent decisions, such as grounding and state checks, to neural models. We learn these policies with neuro-symbolic policy iteration: starting from one agent trajectory, it executes the policy, diagnoses failures with task-completion and step-level judges, and revises the code with a coding model informed by an agent's continuation from the point of failure, without access to the benchmark evaluator. Iterating on generated parameter and initial-state variants makes the policy reusable, and a pre-action verifier guards each state-mutating step at deployment. On OSWorld-Verified and ScienceBoard, the learned policies achieve the highest Pass^3 of all methods in all four settings, 3.6-15.8 points above the base agent, while cutting per-run cost by 15-217$\times$ and latency by 3.4-5.1$\times$. On OSWorld-Verified, policies built only on variants transfer to the held-out original tasks, exceeding AutoRPA by 8.6-17.5 points in Pass^3. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.36927 — 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 2026-10-07T15:54:53.565Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).