Table of Contents: :: Library Catalog

Saved in:

Bibliographic Details
Main Authors:	Xu, Peng, Xiong, Shengwu, Zhang, Jiajun, Chen, Yaxiong, Zhou, Bowen, Loy, Chen Change, Clifton, David A., Lee, Kyoung Mu, Van Gool, Luc, He, Ruiming, Yao, Ruilin, Long, Xinwei, Huang, Jirui, Tian, Kai, Yang, Sa, Shao, Yihua, Feng, Jin, Zhong, Yue, Zhou, Jiakai, Tang, Cheng, Zou, Tianyu, Zhang, Yifang, Liang, Junming, Li, Guoyou, Wang, Zhaoxiang, Zhou, Qiang, Zhao, Yichen, Xiong, Shili, Nam, Hyeongjin, Lee, Jaerin, Chung, Jaeyoung, Park, JoonKyu, Oh, Junghun, Lee, Kanggeon, Lee, Wooseok, Ro, Juneyoung, Osman, Turghun, Hu, Can, Liao, Chaoyang, Chen, Cheng, Han, Chengcheng, Qiu, Chenhao, Peng, Chong, Xu, Cong, Li, Dailin, Wang, Feiyu, Gao, Feng, Zhu, Guibo, Tang, Guopeng, Lu, Haibo, Fang, Han, Qi, Han, Wu, Hanxiao, Cheng, Haobo, Sun, Hongbo, Chen, Hongyao, Hu, Huayong, Li, Hui, Ma, Jiaheng, Yu, Jiang, Wang, Jianing, Yang, Jie, He, Jing, Zhou, Jinglin, Li, Jingxuan, Kittler, Josef, Zheng, Lihao, Zhao, Linnan, Jia, Mengxi, Yan, Muyang, Thien, Nguyen Thanh, Luo, Pu, Li, Qi, Song, Shien, Dong, Shijie, Shao, Shuai, Li, Shutao, Xue, Taofeng, Xu, Tianyang, Gao, Tianyi, Li, Tingting, Zhang, Wei, Su, Weiyang, Dong, Xiaodong, Wu, Xiao-Jun, Zhou, Xiaopeng, Chen, Xin, Wei, Xin, You, Xinyi, Kang, Xudong, Zhou, Xujie, Liu, Xusheng, Wang, Yanan, Huang, Yanbin, Liu, Yang, Yang, Yang, Deng, Yanglin, Kang, Yashu, Yuan, Ye, Wen, Yi, Tian, Yicen, Tao, Yilin, Tang, Yin, Lin, Yipeng, Wang, Yiqing, Xi, Yiting, Yu, Yongkang, Li, Yumei, Qin, Yuxin, Chen, Yuying, Cen, Yuzhe, Zou, Zhaofan, Liu, Zhaohong, Shen, Zhehao, Du, Zhenglin, Li, Zhengyang, Huang, Zhenni, Shao, Zhenwei, Song, Zhilong, Feng, Zhiyong, Wang, Zhiyu, Yu, Zhou, Li, Ziang, Zhai, Zihan, Zhang, Zijian, Peng, Ziyang, Xiao, Ziyun, Li, Zongshu
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition
Online Access:	https://arxiv.org/abs/2509.14142
Tags:	Add Tag No Tags, Be the first to tag this record!

Table of Contents:

This paper reviews the MARS2 2025 Challenge on Multimodal Reasoning. We aim to bring together different approaches in multimodal machine learning and LLMs via a large benchmark. We hope it better allows researchers to follow the state-of-the-art in this very dynamic area. Meanwhile, a growing number of testbeds have boosted the evolution of general-purpose large language models. Thus, this year's MARS2 focuses on real-world and specialized scenarios to broaden the multimodal reasoning applications of MLLMs. Our organizing team released two tailored datasets Lens and AdsQA as test sets, which support general reasoning in 12 daily scenarios and domain-specific reasoning in advertisement videos, respectively. We evaluated 40+ baselines that include both generalist MLLMs and task-specific models, and opened up three competition tracks, i.e., Visual Grounding in Real-world Scenarios (VG-RS), Visual Question Answering with Spatial Awareness (VQA-SA), and Visual Reasoning in Creative Advertisement Videos (VR-Ads). Finally, 76 teams from the renowned academic and industrial institutions have registered and 40+ valid submissions (out of 1200+) have been included in our ranking lists. Our datasets, code sets (40+ baselines and 15+ participants' methods), and rankings are publicly available on the MARS2 workshop website and our GitHub organization page https://github.com/mars2workshop/, where our updates and announcements of upcoming events will be continuously provided.

Similar Items