Y. Qiao, “Aligning Cross-View Visual Geometries in LVLMs Through Human-Like Reasoning Learning”, AAAI, vol. 40, no. 23, pp. 19345–19353, Mar. 2026.