AI RESEARCH
Attribution-Guided Model Rectification of Unreliable Neural Network Behaviors
arXiv CS.AI
•
ArXi:2603.15656v1 Announce Type: cross The performance of neural network models deteriorates due to their unreliable behavior on non-robust features of corrupted samples. Owing to their opaque nature, rectifying models to address this problem often necessitates arduous data cleaning and model re