MARS2 2025 Challenge on Multimodal Reasoning: Datasets, Methods, Results, Discussion, and Outlook

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.

A Behavior-aware Graph Convolution Network Model for Video Recommendation

Interactions between users and videos are the major data source of performing video recommendation. Despite lots of existing recommendation methods, user behaviors on videos, which imply the complex relations between users and videos, are still far from being fully explored. In the paper, we present a model named Sagittarius. Sagittarius adopts a graph convolutional neural network to capture the influence between users and videos. In particular, Sagittarius differentiates between different user behaviors by weighting and fuses the semantics of user behaviors into the embeddings of users and videos. Moreover, Sagittarius combines multiple optimization objectives to learn user and video embeddings and then achieves the video recommendation by the learned user and video embeddings. The experimental results on multiple datasets show that Sagittarius outperforms several state-of-the-art models in terms of recall, unique recall and NDCG.

Improving Sequential Recommendation with Attribute-augmented Graph Neural Networks

Many practical recommender systems provide item recommendation for different users only via mining user-item interactions but totally ignoring the rich attribute information of items that users interact with. In this paper, we propose an attribute-augmented graph neural network model named Murzim. Murzim takes as input the graphs constructed from the user-item interaction sequences and corresponding item attribute sequences. By combining the GNNs with node aggregation and an attention network, Murzim can capture user preference patterns, generate embeddings for user-item interaction sequences, and then generate recommendations through next-item prediction. We conduct extensive experiments on multiple datasets. Experimental results show that Murzim outperforms several state-of-the-art methods in terms of recall and MRR, which illustrates that Murzim can make use of item attribute information to produce better recommendations. At present, Murzim has been deployed in MX Player, one of India's largest streaming platforms, and is recommending videos for tens of thousands of users.