A focused look at this week’s most significant advances in multimodal ai \u2014 6 papers surveyed from arXiv and leading AI labs.<\/p>\n
—<\/p>\n
The boundaries between vision, language, and other modalities continue to blur. This week’s research spans everything from enhanced visual token processing to novel benchmarks for multimodal understanding.<\/p>\n
Unison: Benchmarking Unified Multimodal Models via Synergistic Understanding and Generation<\/strong> \u2014 Jinyu Liu, Xincheng Shuai, Henghui Ding, Yu-Gang Jiang<\/em><\/p>\n Unified multimodal models capable of both understanding and generation have achieved remarkable strides. However, despite their unified designs, existing evaluations typically assess understanding and…<\/p>\n arXiv<\/a><\/p>\n HarmVideoBench: Benchmarking Harmful Video Understanding in Large Multimodal Models<\/strong> \u2014 Jiajun Wu, Haoyu Kang, Yining Sun, Jiacheng Hou, Heng Zhang, Danyang Zhang, Zhenjun Zhao, Haochi Zhang, Leixin Sun, Eric<\/em><\/p>\n Large vision-language models (LVLMs) have recently shown immense potential in automated content moderation, sparking growing interest in developing harmful-video benchmarks. However, we identify two p…<\/p>\n arXiv<\/a><\/p>\n See & Sniff: Learning Visuo-Olfactory Representations<\/strong> \u2014 Seongyu Kim, Seungwoo Lee, Hyeonggon Ryu, Joon Son Chung, Arda Senocak<\/em><\/p>\n While modern multimodal models integrate vision with language, audio, or touch, olfaction remains largely unexplored due to the lack of paired visuo-olfactory data. We introduce SmellNet-V, a scalable…<\/p>\n arXiv<\/a><\/p>\n Paying More Attention to Visual Tokens in Self-Evolving Large Multimodal Models<\/strong><\/p>\n Shravan Venkatraman, Ritesh Thawkar, Omkar Thawakar, Rao Muhammad Anwer, Hisham Cholakkal, Salman Kh<\/em><\/p>\n Recently, self-evolving large multimodal models (LMMs) have received attention for improving visual reasoning in a purely unsupervised setting. However, multi-role self-play and self-consistency reward schemes in existing self-evolving LMMs…<\/p>\n arXiv<\/a><\/p>\n Risk-Aware Selective Multimodal Driver Monitoring with Driver-State World Modeling<\/strong><\/p>\n Daosheng Qiu, Haozhuang Chi, Hao Su, Shu Long, Xinyue Miao, Yongle Dong, Wei Zhang<\/em><\/p>\n Continuous driver monitoring in automated vehicles requires low-latency inference while avoiding unsafe decisions under uncertain driver states. Large vision-language models provide broad multimodal priors, but their latency and limited reliability…<\/p>\nReasoning & Inference<\/h2>\n
Multimodal<\/h2>\n