Metraipolis › archive › dc_28151c705227

UI-Venus-2 Technical Report — arXiv:2609.00028

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.00028
License (as archived)
CC0-1.0
Archived
2026-10-17T06:31:37.983Z
SHA-256 of the archived text
a8ef93881033fae16c71dfc521e96b4a2bd7feac746848cd516265012f9072dd
Stable URL of this page
https://metraipolis.com/documento/dc_28151c705227

How to cite

UI-Venus-2 Technical Report — arXiv:2609.00028 — arXiv (en), https://arxiv.org/abs/2609.00028 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_28151c705227 (Metraipolis archive dc_28151c705227, sha256 a8ef93881033fae16c71dfc521e96b4a2bd7feac746848cd516265012f9072dd)
BibTeX
@misc{dc28151c705227,
  title = {UI-Venus-2 Technical Report — arXiv:2609.00028},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.00028},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_28151c705227; sha256 a8ef93881033fae16c71dfc521e96b4a2bd7feac746848cd516265012f9072dd},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.00028. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multimodal GUI agents have emerged as a promising paradigm for digital task automation, yet transitioning from benchmark-oriented models to dependable real-world applications remains challenging due to limited environment coverage, brittle task construction, and unreliable reward verification. In this work, we present UI-Venus-2, a general-purpose foundation GUI agent designed to operate across mobile, web, and desktop environments through a unified closed-loop reasoning-action framework. To bridge the gap toward practical deployment, we jointly scale three critical dimensions: (1) Environments, expanding coverage to more than 170 multilingual mobile apps and native desktop operating systems; (2) Tasks, employing a deep-research pipeline for function-grounded instruction generation; and (3) Verification, adopting trace-level and sample-level evaluators with visual keypoints and multi-model voting to ensure reliable RL signals for training. Furthermore, we integrate safety-aware mechanisms to ensure controlled execution of consequential actions. By offering a capable, efficient, and open-source foundation, UI-Venus-2 advances the field toward more generalizable, verifiable, and self-reflective agents for real-world applications.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.00028 — 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 2026-10-17T06:15:12.547Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).