Metraipolis › archive › dc_3fd89e406ea7
InterEvolve: Test-Time Evolution of Reward Programs for Humanoid Loco-Manipulation — arXiv:2610.02196
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.02196
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T19:17:01.020Z
- SHA-256 of the archived text
3efed5bdc07ae27a657005fa669229985d2604fea42a8c75aabdd67dd22a68de- Stable URL of this page
https://metraipolis.com/documento/dc_3fd89e406ea7
How to cite
InterEvolve: Test-Time Evolution of Reward Programs for Humanoid Loco-Manipulation — arXiv:2610.02196 — arXiv (en), https://arxiv.org/abs/2610.02196 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3fd89e406ea7 (Metraipolis archive dc_3fd89e406ea7, sha256 3efed5bdc07ae27a657005fa669229985d2604fea42a8c75aabdd67dd22a68de)
BibTeX
@misc{dc3fd89e406ea7,
title = {InterEvolve: Test-Time Evolution of Reward Programs for Humanoid Loco-Manipulation — arXiv:2610.02196},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.02196},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3fd89e406ea7; sha256 3efed5bdc07ae27a657005fa669229985d2604fea42a8c75aabdd67dd22a68de},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.02196. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We study test-time evolution for humanoid loco-manipulation: solving tasks that a controller was never trained for by repurposing its existing skills, improving from its own attempts, and retaining what it learns, without retraining. Our key insight is that a broad controller already holds much of the competence a new task needs, and that this competence becomes accessible through an interface between planning and control that is expressive enough to specify contact-rich, multi-stage interactions, yet executable and measurable enough that execution feedback can guide planning from experience. InterEvolve realizes this interface with two components. First, we develop an object-aware forward-backward (FB) behavioral foundation model, whose object residuals on a frozen body prior turn a new reward about the body or objects into loco-manipulation behavior at test time. Second, we specify tasks as reward programs: staged rewards with completion conditions and tunable constants. A large language model (LLM) agent revises the program structure in context, drawing on execution feedback and a skill library of verified programs, while a numerical optimizer tunes its constants. With every candidate verified across parallel simulation scenarios, the program explores new ways to induce, repurpose, and compose the controller's existing motor competence for the task at hand, and thus improves over iterations. Experiments show that human-designed rewards leave much of the FB model's loco-manipulation competence untapped, whereas the programs InterEvolve evolves release it, sometimes through novel strategies. It further produces behaviors for diverse tasks, complex scenes, and long-horizon compositions in simulation, and evolved skills run autonomously on a physical Unitree G1 from egocentric onboard perception. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CV). Categorie arXiv: Robotics (cs.RO) ; Computer Vision and Pattern Recognition (cs.CV); Graphics (cs.GR). Pagina del lavoro: https://arxiv.org/abs/2610.02196 — 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 2026-10-06T17:37:36.961Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).