No article text was recorded for this version.
internal activation monitoring
Version 1 of 1191 papers249 conceptsfrozen 2026-10-04
Papers in this version
- Blog: SafetyNet: Detecting and Rejecting Adversarial Examples Robustly
- Blog: Detecting Adversarial Perturbations Through Spatial Behavior in Activation Spaces
- Blog: DAAIN: Detection of Anomalous and Adversarial Input using Normalizing Flows
- Blog: What Models Know About Their Attackers: Deriving Attacker Information From Latent Representations
- Blog: Unravelling Robustness of Deep Learning Based Face Recognition Against Adversarial Attacks
- Blog: Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering
- Blog: NIC: Detecting Adversarial Samples with Neural Network Invariant Checking
- Blog: Effective and Robust Detection of Adversarial Examples via Benford-Fourier Coefficients
- Blog: Mitigating Deep Reinforcement Learning Backdoors in the Neural Activation Space
- Blog: Hidden in Memory: Sleeper Memory Poisoning in LLM Agents
- Blog: SafetyNet: Detecting and Rejecting Adversarial Examples Robustly
- Blog: FedDebug: Systematic Debugging for Federated Learning Applications
- Blog: Interpretable Neural Network Decoupling
- Blog: Attack Agnostic Statistical Method for Adversarial Detection
- Blog: Prompt Perturbation in Retrieval-Augmented Generation based Large Language Models
- Blog: Ptolemy: Architecture Support for Robust Deep Learning
- Blog: Quality Assurance of A GPT-based Sentiment Analysis System: Adversarial Review Data Generation and Detection
- Blog: Hidden State Poisoning Attacks against Mamba-based Language Models
- Blog: Spectral Signatures in Backdoor Attacks
- Blog: A Sweet Rabbit Hole by DARCY: Using Honeypots to Detect Universal Trigger's Adversarial Attacks
- Blog: Detection of Face Recognition Adversarial Attacks
- Blog: Anomalous Example Detection in Deep Learning: A Survey
- Blog: SAFE-NID: Self-Attention with Normalizing-Flow Encodings for Network Intrusion Detection
- Blog: Gotta Catch 'Em All: Using Concealed Trapdoors to Detect Adversarial Attacks on Neural Networks
- Blog: The Taboo Trap: Behavioural Detection of Adversarial Samples
- Blog: Defending from Physically-Realizable Adversarial Attacks through Internal Over-Activation Analysis
- Blog: Out of Distribution Detection and Adversarial Attacks on Deep Neural Networks for Robust Medical Image Analysis
- Blog: Mitigating Deep Reinforcement Learning Backdoors in the Neural Activation Space
- Blog: Testing the Channels of Convolutional Neural Networks
- Blog: Identifying Adversarial Attacks on Text Classifiers
- Blog: Runtime Safety Monitoring of Deep Neural Networks for Perception: A Survey
- Blog: NeuralDivergence: Exploring and Understanding Neural Networks by Comparing Activation Distributions
- Blog: Interpret Neural Networks by Identifying Critical Data Routing Paths
- Blog: Hamming Similarity and Graph Laplacians for Class Partitioning and Adversarial Image Detection
- Blog: Data Sanity Check for Deep Learning Systems via Learnt Assertions
- Blog: FedDefender: Backdoor Attack Defense in Federated Learning
- Blog: Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Blog: ControlNET: A Firewall for RAG-based LLM System
- Blog: Kill Two Birds with One Stone! Trajectory enabled Unified Online Detection of Adversarial Examples and Backdoor Attacks
- Blog: Towards Creativity Characterization of Generative Models via Group-based Subset Scanning
- Blog: CatchBackdoor: Backdoor Detection via Critical Trojan Neural Path Fuzzing
- Blog: NoiseCAM: Explainable AI for the Boundary Between Noise and Adversarial Attacks
- Blog: A Novel Statistical Measure for Out-of-Distribution Detection in Data Quality Assurance
- Blog: Can We Mitigate Backdoor Attack Using Adversarial Detection Methods?
- Blog: Towards AI Forensics: Did the Artificial Intelligence System Do It? Why?
- Blog: Know Your Enemy: Identifying Adversarial Behaviours in Deep Reinforcement Learning Agents (Student Abstract)
- Blog: Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering
- Blog: Detecting Attacks Against Deep Reinforcement Learning for Autonomous Driving
- Blog: Adversarial Example Detection in Deployed Tree Ensembles
- Blog: Robust Detection of Adversarial Attacks by Modeling the Intrinsic Properties of Deep Neural Networks
- Blog: MDPFuzz: testing models solving Markov decision processes
- Blog: Self-Ensembling Vision Transformer (SEViT) for Robust Medical Image Classification
- Blog: Statistical Process Monitoring of Artificial Neural Networks
- Blog: TEN-GUARD: Tensor Decomposition for Backdoor Attack Detection in Deep Neural Networks
- Blog: Testing the Channels of Convolutional Neural Networks
- Blog: GAN Ensemble for Anomaly Detection
- Blog: A Unified Detection Framework for Inference-Stage Backdoor Defenses
- Blog: Detecting Adversarial Attacks via Subset Scanning of Autoencoder Activations and Reconstruction Error
- Blog: Outlier Detection through Null Space Analysis of Neural Networks
- Blog: HAWKEYE: Adversarial Example Detector for Deep Neural Networks
Showing 60 of 191 papers.
History & citation
Current · Permanent link
- This direction split from Mechanistic Anomaly Detection via Internal Activation Trajectories (53% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_internal_activation_monitoring_f75773bc,
title = {internal activation monitoring},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/internal-activation-monitoring/rdr_f75773bc27b244edb0bb5fa4360cdad4},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}This exact version (version 1, frozen text):
@misc{lacuna_internal_activation_monitoring_f75773bc_v1,
title = {internal activation monitoring},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/internal-activation-monitoring/rdr_f75773bc27b244edb0bb5fa4360cdad4/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}