Add a new reasoning benchmark MM-IQ

[AceCHQ]

Thank you for your excellent job. We want to add a new systematic benchmark MM-IQ for LMM to your repo. Thank you.

Paper: https://arxiv.org/pdf/2502.00698 Code: https://github.com/AceCHQ/MMIQ/tree/main Project: https://acechq.github.io/MMIQ-benchmark/ Leaderboard: https://acechq.github.io/MMIQ-benchmark/#leaderboard

Citation: @misc{cai2025mm-iq, title={MM-IQ: Benchmarking Human-Like Abstraction and Reasoning in Multimodal Models}, author={Huanqia Cai and Yijun Yang and Winston Hu}, journal={arXiv preprint arXiv:2502.00698}, year={2025} }