Islam, Md Mofijul, et al. “PATRON: Perspective-Aware Multitask Model for Referring Expression Grounding Using Embodied Multimodal Cues”. Proceedings of the AAAI Conference on Artificial Intelligence, vol. 37, no. 1, June 2023, pp. 971-9, doi:10.1609/aaai.v37i1.25177.