A. B. M. Ashikur Rahman Saeed Anwar Muhammad Usman 0010 Irfan Ahmad 0001 Ajmal Mian PENDULUM: A Benchmark for Assessing Sycophancy in Multimodal Large Language Models. 2025 December abs/2512.19350 CoRR https://doi.org/10.48550/arXiv.2512.19350 db/journals/corr/corr2512.html#abs-2512-19350 streams/journals/corr