Talks

Invited

2025

Amazon Nova AI Challenge 2025
PurpCode - The First Reasoning Model for Cybersafety

June 05, 2025

Slides
UIUC 2025
Can LLMs identify Security Vulnerabilities?

April 17, 2025

Slides

2024

UIUC 2024
Jailbreaking of LLMs

October 10, 2024

Slides
UIUC 2024
Generating Sequences by Learning to Self-Correct

March 06, 2024

Slides

2023

UIUC 2023
Promptly Racing Ahead - A Survey on Multi-task Prompt-Based Learning

May 01, 2023

Slides
UIUC 2023
Editing Models with Task Arithmetic

April 01, 2023

Slides

2022

UIUC 2022
A Pseudo-Labelling Approach for Low-resource Domain Adaptation

August 01, 2022

Slides ★

2021

Security Machine Learning Seminar 2021
Weight Poisoning Attack on Pre-trained Language Model

June 01, 2021

Slides
Security Machine Learning Seminar 2021
Practical No-box Adversarial Attacks against DNNs

May 01, 2021

Slides