# Utilizing Network Properties to Detect Erroneous Inputs

arXiv: https://arxiv.org/abs/2002.12520 · HTML: https://mlsanity.com/paper/2002.12520/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Feb 2020.

## Built on by (most cited first, 7 of 7)

- [Out-of-Distribution Data: An Acquaintance of Adversarial Examples -- A Survey](https://mlsanity.com/paper/2404.05219/) (2024)
- [Cross-Model Disagreement as a Label-Free Correctness Signal](https://mlsanity.com/paper/2603.25450/) (2026)
- [Hamming Similarity and Graph Laplacians for Class Partitioning and Adversarial Image Detection](https://mlsanity.com/paper/2305.01808/) (2023)
- [Dual Graphs of Polyhedral Decompositions for the Detection of Adversarial Attacks](https://mlsanity.com/paper/2211.13305/) (2022)
- [Cross-Silo Federated Learning Across Divergent Domains with Iterative Parameter Alignment](https://mlsanity.com/paper/2311.04818/) (2023)
- [Tiled Bit Networks: Sub-Bit Neural Network Compression Through Reuse of Learnable Binary Vectors](https://mlsanity.com/paper/2407.12075/) (2024)
- [FeatureLens: A Highly Generalizable and Interpretable Framework for Detecting Adversarial Examples Based on Image Features](https://mlsanity.com/paper/2512.03625/) (2025)

## Abstract (arXiv)

Neural networks are vulnerable to a wide range of erroneous inputs such as adversarial, corrupted, out-of-distribution, and misclassified examples. In this work, we train a linear SVM classifier to detect these four types of erroneous data using hidden and softmax feature vectors of pre-trained neural networks. Our results indicate that these faulty data types generally exhibit linearly separable activation properties from correct examples, giving us the ability to reject bad inputs with no extra training or overhead. We experimentally validate our findings across a diverse range of datasets, domains, pre-trained models, and adversarial attacks.