Data Poisoning Discussion

​This week we are going over data poisoning. This is when "malicious data is used to corrupt or manipulate the training data of AI and ML models". This is a topic that anyone passionate about AI should know about! Meet us at Harold Library to go over the following article.

​​Paper: "Data Poisoning in Deep Learning: A Survey"

​Paper: 2503.22759

​Respository of the paper: Pinlong-Zhao/Data-Poisoning: Data Poisoning in Deep Learning: A Survey

​

​Extra Credit:

​Paper: Not What You've Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection

​Repository : tamlhp/awesome-recsys-poisoning: A Survey of Poisoning Attacks and Defenses in Recommender Systems

​​Meeting is going to be at:
5th Fl. North - Harold Washington Library Center