@inproceedings{liang-etal-2026-capmem, title = "{C}ap{M}em: A Benchmark for Caption-Based Episodic Memory in Egocentric Video", author = "Liang, Dingli and Xie, Yiqiao and Huang, Yukai and Wang, Zhaokai and Cai, Weitong and Feng, Guangwen and Song, Jifei and Zhang, Zhensong and Zhang, Hang", editor = "Atanasova, Pepa and Haffari, Gholamreza and Wang, Minghan", booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: Industry Track", month = oct, year = "2026", address = "Budapest, Hungary", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2026.emnlp-industry.233/", pages = "3888--3914", ISBN = "979-8-89176-460-6" }