arXiv:2608.29841v1 Announce Type: new Abstract: Generating formally verified programs from natural language remains challenging: existing approaches either produce code in a single pass without recour...
arXiv:2608.26716v1 Анонс Тип: новый Абстракт: понимание планировки или интерпретация организации элементов имеет важное значение для анализа документов, создания пользовательского интерфейса (UI) и гр...
arXiv:2608.21993v1 Announce Type: new Abstract: Мы представляем Gated Decoupled Compositional Bandits (GDCB), семейство контекстных бандитских алгоритмов с тремя структурными инновациями, которые совм...
arXiv:2608.18171v1 Новые модели языка Looped показали многообещающие результаты на тестах рассуждений, но их потенциал для использования агентных инструментов остается в значительной степени неисследо...
arXiv:2608.19871v1 Announce Type: new Abstract: Compositional Zero-Shot Learning (CZSL) aims to recognize unseen attribute-object compositions by leveraging knowledge of primitive concepts learned fro...
arXiv:2608.17351v1 Announce Type: new Abstract: Open-world face anti-spoofing must address both covariate and semantic shifts: source and target domains differ in imaging conditions, while target doma...
arXiv:2608.14741v1 Announce Type: new Abstract: We introduce PolyComp, a procedurally generated and verified benchmark that stresses visual recognition and compositional spatial reasoning. In each pro...
arXiv:2512.23754v2 Announce Type: replace Abstract: Building multiscale biological models requires the integration of independently developed submodels, which means moving shared variables between the...
arXiv:2608.12895v1 Announce Type: new Abstract: Compositional reliability bounds for multi-agent systems multiply component reliabilities, a step licensed by a conditional-independence assumption that...