作者位次 #3/7ICML 2026Poster
Calibrated Multimodal Representation Learning with Missing Modalities
Xiaohao Liu · Xiaobo Xia · Jiaheng Wei · Shuo Yang · Xiu Su · See-Kiong Ng · Tat-Seng Chua
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Xiaohao Liu · Xiaobo Xia · Jiaheng Wei · Shuo Yang · Xiu Su · See-Kiong Ng · Tat-Seng Chua
Na Di · Ling Li · Zhe Tang · Hao Cheng · Jinlong Pang · Jiaheng Wei · Zhaowei Zhu
Zimu Jia · Mingjie Xu · Andrew Estornell · Jiaheng Wei
The efficacy of Large Vision-Language Models (LVLMs) is critically dependent on the quality of their training data, requiring a precise balance between visual fidelity and instruction-following capability. Existing datasets, however, are plagued by inconsistent quality, and current data filtering...
Hao Zheng · Zirui Pang · Ling Li · Zhijie Deng · Yuhan Pu · Zhaowei Zhu · Xiaobo Xia · Jiaheng Wei
Advances in Multimodal Large Language Models (MLLMs) intensify concerns about data safety, making Machine Unlearning (MU), the selective removal of harmful/private information, a critical necessity. However, existing MU benchmarks for MLLMs are limited by a lack of image diversity, coarse-grained...
Yanshan Liu · Hongbo Zhang · Zhen Sun · Jiaheng Wei · Kaishun Wu
In complex domains like interior design, user requests are often ambiguous and multimodal. Professional designers address this by asking strategic clarification questions based on hierarchical priorities, a capability lacking in current Vision-Language Models (VLMs). When fine-tuned on dialogue d...
Yanji He · Yuxin Jiang · Yiwen Wu · Bo Huang · Jiaheng Wei · Wei Wang
Large Language Models are increasingly deployed for decision-making, yet their adoption in high-stakes domains remains limited by miscalibrated probabilities, unfaithful explanations, and inability to incorporate expert knowledge precisely. We propose IDEA , a framework that extracts LLM decision...