Prior work on probing neural networks primarily relies on input-space analysis or parameter perturbation, both of which face fundamental limitations in accessing structural information encoded in intermediate representations. We introduce Activation Perturbation for EXploration (APEX), an inference-time probing paradigm that perturbs hidden activations while keeping both inputs and model parameters fixed. We theoretically show that activation perturbation induces a principled transition from sample-dependent to model-dependent behavior by suppressing input-specific signals and amplifying representation-level structure, and further establish that input perturbation corresponds to a constrained special case of this framework. Through representative case studies, we demonstrate the practical advantages of APEX. In the small-noise regime, APEX provides a lightweight and efficient measure of sample regularity that aligns with established metrics, while also distinguishing structured from randomly labeled models and revealing semantically coherent prediction transitions. In the large-noise regime, APEX exposes training-induced model-level biases, including a pronounced concentration of predictions on the target class in backdoored models. Overall, our results show that APEX offers an effective perspective for exploring, and understanding neural networks beyond what is accessible from input space alone.
Probing Graph Neural Network Activation Patterns Through Graph Topology
Department of Computer Science, University of Geneva, Geneva, Switzerland（日内瓦大学计算机科学系）；Research Institute for Statistics and Information Science, Geneva, Switzerland（日内瓦统计与信息科学研究所）；School of Engineering and Applied Sciences, Harvard University（哈佛大学工程与应用科学学院）
Detection Without Correction: A Robust Asymmetry in Activation-Based Hallucination Probing
Department of Computer Science and Engineering, Indian Institute of Technology Patna, India（印度理工学院帕纳分校计算机科学与工程系）；Department of Computer Science, Rajarshi School of Management and Technology, Varanasi, India（印度瓦拉纳西 Rajarshi 管理与技术学校计算机科学系）；Department of Electronics and Communication Engineering, Jaypee Institute of Information Technology, Noida, India（印度诺伊达 Jaypee 信息科技学院电子与通信工程系）
Probing the Robustness of Large Language Models Safety to Latent Perturbations
Tsinghua Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际 Graduate School, 清华大学）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；Fudan University（复旦大学）；The University of Hong Kong（香港大学）
Stanford University（斯坦福大学）；Polytechnic University of Catalonia（加泰罗尼亚理工大学）；Georgia Institute of Technology（佐治亚理工学院）；Skoltech（斯科尔泰克科技大学）；University of Queensland（昆士兰大学）；UC Berkeley（加州大学伯克利分校）；Alignment Research Center（对齐研究中心）；MIT CSAIL（麻省理工学院计算机科学与人工智能实验室）；Chan Zuckerberg Biohub（陈·扎克翰生物中心）
Probing Quantum Spin Systems with Kolmogorov-Arnold Neural Network Quantum States
Department of Physics and Astronomy, University of Alabama（物理与天文学系，阿拉巴马大学）；Department of Physics and Astronomy, University of Kansas（物理与天文学系，堪萨斯大学）；Department of Physics, University of Dhaka（物理系，达卡大学）
