OFFSIDE: Benchmarking Unlearning Misinformation in Multimodal Large Language Models Paper • 2510.22535 • Published Oct 26, 2025 • 1
