Metraipolis › archive › dc_af595f30980a

RoboFoundry: System-as-Policy Evolution for Self-Learning Embodied Agents — arXiv:2609.32862

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.32862
License (as archived)
CC0-1.0
Archived
2026-10-07T21:00:42.545Z
SHA-256 of the archived text
8b07e78a09a90e0046f1e83e4633af4fd6a7b5f42d0f8d0ad5f734ca1bd11c2e
Stable URL of this page
https://metraipolis.com/documento/dc_af595f30980a

How to cite

RoboFoundry: System-as-Policy Evolution for Self-Learning Embodied Agents — arXiv:2609.32862 — arXiv (en), https://arxiv.org/abs/2609.32862 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_af595f30980a (Metraipolis archive dc_af595f30980a, sha256 8b07e78a09a90e0046f1e83e4633af4fd6a7b5f42d0f8d0ad5f734ca1bd11c2e)
BibTeX
@misc{dcaf595f30980a,
  title = {RoboFoundry: System-as-Policy Evolution for Self-Learning Embodied Agents — arXiv:2609.32862},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.32862},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_af595f30980a; sha256 8b07e78a09a90e0046f1e83e4633af4fd6a7b5f42d0f8d0ad5f734ca1bd11c2e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32862. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A foundation model should not act in isolation as an embodied agent. Yet, existing methods often optimize individual components of the agent stack, such as memory, context, skills, or action interfaces, rather than treating the supporting system itself as a unified policy. Moreover, interaction alone does not yield self-improvement unless execution experience is converted into persistent, validated system changes. We therefore propose RoboFoundry, the first embodied agentic framework that formulates this process as Self-Evolving System-as-Policy. RoboFoundry diagnoses capability gaps in decision-making and memory management, converts execution traces into validated task-specific system updates, and promotes recurring improvements to the general system. Evolution operates over two complementary surfaces: a context system that manages active internal context and persistent file-system memory, and a hierarchical skill system that organizes atomic skills, reusable compositions, and failure-conditioned recovery. A shared semantic interface separates embodiment-invariant decisions from embodiment-specific execution, allowing evolved system capabilities to transfer across heterogeneous robots. On EmbodiedBench, RoboFoundry achieves state-of-the-art performance, notably improving GPT-5.5 by 27.8%. It also brings Qwen3.7-Plus to near parity with GPT-5.5 (70.3% vs. 72.7%), showing consistent gains from system-as-policy evolution across foundation models. For long-horizon memory, RoboFoundry outperforms all baselines on RoboMemArena by at least 39.0%, even against methods assisted by external foundation models. On LIBERO-PRO, it further outperforms Cap-Agent0 by 243.8%-679.7% across all perturbation types. In real-world deployments, RoboFoundry demonstrates zero-shot transfer and online evolution across robots and tasks, highlighting its potential for fully autonomous embodied agents.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.32862 — 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 2026-10-07T20:29:02.482Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).