@inproceedings{radwan-etal-2026-sonic, title = "{SONIC}-{O}1: A Real-World Benchmark for Evaluating Multimodal Large Language Models on Audio-Video Understanding", author = "Radwan, Ahmed Y. and Emmanouilidis, Christos and Tabassum, Hina and Pandya, Deval and Raza, Shaina", editor = "Lee, Hung-Yi and Dev, Sunipa and Li, Jessy and Elliott, Desmond", booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing", month = oct, year = "2026", address = "Budapest, Hungary", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2026.emnlp-main.767/", pages = "17786--17825", ISBN = "979-8-89176-459-0" }