Completed
Intro
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
PICCOLO - Exposing Complex Backdoors in NLP Transformer Models
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Backdoor Attacks on NLP Models
- 3 Trigger Inversion is highly effective at detecting backdoors in Computer Vision
- 4 Structure of NLP Transformers
- 5 Challenge I: Input Space is Discrete, and NLP Models are not Differentiable to Input
- 6 Proposal to Challenge I: Differentiable Model Transformation
- 7 Challenge II: Token Level Optimization cannot Reverse Engineer Complex Words with Multiple Tokens.
- 8 Proposal to Challenge II: Word Level Inversion
- 9 Overview
- 10 Evaluation Setup
- 11 Effectiveness
- 12 Code Repo
