Issues

• Switch to HuggingFace for Datasets in Examples

  pralab/secml-torch#209 · Filter by author fabiobrau opened on Jul 31, 2026
  1

• Trackers receive the gradient from the previous iteration

  pralab/secml-torch#207 · Filter by author cossu555 opened on Jul 15, 2026

• Simplify imports

  enhancementrefactoring
  pralab/secml-torch#206 · Filter by author maurapintor opened on Jun 17, 2026
  

• QuantizationConstraint class - out of memory error

  pralab/secml-torch#202 · Filter by author pintorem opened on May 25, 2026
  2

  

• Device mismatch in ModularEvasionAttackMinDistance

  bug
  Bug

  pralab/secml-torch#198 · Filter by author maurapintor opened on Apr 20, 2026
  1

  

• Black-box optimization

  help wanted
  Feature

  pralab/secml-torch#197 · Filter by author maurapintor opened on Apr 15, 2026

• Method name conflict: BasePyTorchClassifier.train overlaps with nn.Module.train semantics

  refactoring
  pralab/secml-torch#185 · Filter by author fabiobrau opened on Mar 25, 2026
  1

  

• Add HarmBench Jailbreaks wrapper

  enhancement
  pralab/secml-torch#178 · Filter by author rmura498 opened on Jan 28, 2026
  

• Add object detection task

  enhancementgood first issuehelp wanted
  Feature

  pralab/secml-torch#176 · Filter by author maurapintor opened on Jan 23, 2026
  2

• Add privacy attacks

  help wanted
  Feature

  pralab/secml-torch#175 · Filter by author maurapintor opened on Jan 21, 2026
  

• Native evasion attacks keep dataloader on CPU when model is on CUDA

  bug
  pralab/secml-torch#173 · Filter by author lucascionis opened on Nov 24, 2025
  1

  

• BaseJailbreak — skeleton for text-attack interface

  enhancement
  pralab/secml-torch#169 · Filter by author rmura498 opened on Nov 11, 2025
  

• Implement of robust trainer against backdoor attack

  pralab/secml-torch#167 · Filter by author guowei-cn opened on Oct 29, 2025

• Model Ensemble transfer attacks

  documentationenhancement
  pralab/secml-torch#163 · Filter by author maurapintor opened on Oct 7, 2025
  

• Add APGD (possibly wrapped and native)

  enhancementhelp wanted
  pralab/secml-torch#159 · Filter by author maurapintor opened on Oct 6, 2025
  1

  

• Add robustness curve utilities for min-norm attacks

  enhancement
  Feature

  pralab/secml-torch#156 · Filter by author lucascionis opened on Oct 3, 2025
  

• Tutorial: wrapping attacks from other libraries

  documentation
  pralab/secml-torch#150 · Filter by author maurapintor opened on Sep 15, 2025
  

• Implement Universal attacks

  enhancement
  pralab/secml-torch#95 · Filter by author maurapintor opened on Oct 4, 2024