Zhang, Yipeng, et al. “LLaVA-UHD v2: Exploiting Hierarchical Vision Granularity in MLLMs via Inverse Semantic Pyramid”. Proceedings of the AAAI Conference on Artificial Intelligence, vol. 40, no. 15, Mar. 2026, pp. 12934-42, doi:10.1609/aaai.v40i15.38292.