mlsanityData to 14 Oct 2026

ATLAS-AL: Adaptive Trust-Region for Latent Adversarial Searches via Active Learning

arXiv 2610.07323 ↗First posted 5 Oct 2026cs.LG cs.CR cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Security evaluation of learning-based systems requires more than just testing the system against a fixed collection of attacks. It requires adaptive mechanisms that can efficiently discover textitsets of inputs that induce model failure. We introduce ATLAS (Adaptive Trust-Regions for Latent Adversarial Searches), which is a query-based framework that discovers adversarial input sets for black-box learning systems. ATLAS casts attack generation as an active learning level set estimation problem then combines calibrated approximations with a local-global sampling architecture to find regions of the input space that contain adversarial examples. Once discovered, ATLAS is designed to sample points within these adversarial regions to build adversarial sets that accurately represent the state of robustness of the target model. When applied on toy experiments, we find that ATLAS is able to recover more of the adversarial region under a limited query budget than does previous work. When applied to standard and adversarially trained MNIST, CIFAR, and ImageNet model targets, ATLAS produces better representative attacks than other query-based black-box attacks (NES, SignHunter, BayesOpt). ATLAS represents an automated red-teaming framework that can be used for both analyzing the robustness of learning-based systems under development and continuous auditing to see how the robustness of a system changes over time.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 11 of its 43 references that are in the map. First posted on arXiv 5 Oct 2026.

Builds on

11 of 43 references.

References · 11

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  2. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
  3. 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
  4. 2016Adversarial Machine Learning at Scalecited by 936
  5. 2018Generating Adversarial Examples with Adversarial Networkscited by 258
  6. 2021Fixing Data Augmentation to Improve Adversarial Robustnesscited by 204
  7. 2025The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injectionscited by 129
  8. 2019Black-box Adversarial Attacks with Bayesian Optimizationcited by 13
  9. 2024Efficient Black-box Adversarial Attacks via Bayesian Optimization Guided by a Function Priorcited by 11
  10. 2021Multi-Adversarial Safety Analysis for Autonomous Vehiclescited by 2
  11. 2026High-dimensional Level Set Estimation with Trust Regions and Double Acquisition Functionscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Adversarial Agents: Black-Box Evasion Attacks with Reinforcement Learning
  2. 2024Defending Against Unforeseen Failure Modes with Latent Adversarial Training
  3. 2026Adversarial Robustness of Activation Steering in Large Language Models
  4. 2026WMAttack: Automated Attack Search for Adversarial Evaluation of World-Model Agents
  5. 2024Steering Away from Harm: An Adaptive Approach to Defending Vision Language Model Against Jailbreaks
  6. 2026Latent Geometric Chords for Query-Efficient Decision-Based Adversarial Attacks
  7. 2025Sampling-aware Adversarial Attacks Against Large Language Models
  8. 2026Risk Under Pressure: Compute-Aware Evaluation of Adversarial Robustness in Language Models