Ma, J. et al. 2026. MMG-Vid: Maximizing Marginal Gains at Segment-level and Token-level for Efficient Video LLMs. Proceedings of the AAAI Conference on Artificial Intelligence. 40, 29 (Mar. 2026), 24253–24261. DOI:https://doi.org/10.1609/aaai.v40i29.39605.