作者位次 #2/5ECCV 2026PosterMitigating Positional Leakage in 3D Masked Autoencoders for Robust Representation LearningXu Yan · Huiqun Wang · Chen Wang · Lei Ren · Di Huang可解释性长上下文与架构3D Reconstruction, Registration and Scene ModelingOpenReview
作者位次 #2/4CVPR 2026Poster· Computer Vision & Pattern RecognitionCoVFT: Context-aware Visual Fine-tuning for Multimodal Large Language ModelsNan Zhou · Huiqun Wang · Yaoyan Zheng · Di Huang多模态强化学习与后训练OpenReview 项目页
作者位次 #2/3CVPR 2026Poster· Computer Vision & Pattern RecognitionEgoMind: Activating Spatial Cognition through Linguistic Reasoning in MLLMsZhenghao Chen · Huiqun Wang · Di Huang推理多模态多语言与语言学OpenReview 项目页