phenomenon · first seen 2 Nov 2026
adversarial attack on text classifiers
Also: text adversarial attack
Crafting input text perturbations that steer a model's prediction toward an attacker-chosen label.
Mentions per week
PapersCodeDiscussion
Earliest sources
- JevBreak: Your Model, Their DecisionPapers · 2 Nov 2026 · Nitin Aravind Birur, Divyanshu Kumar, Sahil Agarwal, Prashanth Harshangi
Latest sources
- JevBreak: Your Model, Their DecisionPapers · 2 Nov 2026 · Nitin Aravind Birur, Divyanshu Kumar, Sahil Agarwal, Prashanth Harshangi