ICON: Invariant Counterfactual Optimization with Neuro-Symbolic Priors for Text-Based Person Search
ICON: 基于神经符号先验的文本基人物搜索不变反事实优化
机构 * College of Computer Science and Engineering, Northeastern University(东北大学计算机科学与工程学院) ; Key Laboratory of Intelligent Computing in Medical Image of Ministry of Education, Northeastern University(教育部智能医学图像计算重点实验室,东北大学) ; School of AI, Beijing Institute of Technology(北京理工大学人工智能学院) ; Foshan Graduate School of Innovation, Northeastern University(创新研究生院,东北大学)
AI总结 ICON通过引入神经符号先验,结合因果和拓扑先验,解决文本基人物搜索中的鲁棒性问题,提升对遮挡、背景干扰和定位噪声的抗性。