NeurIPS

The Unseen Hand: Manipulating Model Fairness and SHAP with Targeted Identity Re-Association Attacks

无形之手：通过目标身份重新关联攻击操纵模型公平性和SHAP

Sannaan Khan, Muhammad U. S. Khan

机构 * National University of Sciences and Technology (NUST)（国立科技大学（NUST）） 纠错

AI总结 提出目标身份重新关联（TIRA）攻击，通过概率性微扰操纵模型输出，在不留痕迹的情况下扭曲公平性指标和SHAP解释。

Comments Accepted at NeurIPS Workshops 2025

Dual-Stream EEG Decoding for 3D Visual Perception

用于3D视觉感知的双流脑电解码

Ninon Lizé Masclef, Taisija Demcenko, Antonella Catanzaro, Nataliya Kosmyna

机构 * Massachusetts Institute of Technology（麻省理工学院） 纠错

AI总结 提出一种模仿生物视觉腹侧和背侧通路的双流脑电解码模型，通过圆形回归预测角度和EEG条件多视图扩散实现3D重建，揭示了时间动态的通道参与模式。

Comments 17 pages, 4 figures. Accepted at the Symmetry and Geometry in Neural Representations Workshop (NeurReps), NeurIPS 2025. To appear in Proceedings of Machine Learning Research (PMLR)