DetectiveSAM: AI Edits Localization
AI & ML interests
Research Demos and Tools for Trustworthy and Safe AI Development and Deployment
Recent Activity
Attention Tracker Demo for Prompt Injection Detection
LLM LabSafety Benchmark
Beyond Demo
Adversarial Example Detector published at ICML 2024: https://proceedings.mlr.press/v235/he24l.html
DetectiveSAM: AI Edits Localization
DivEye: Diversity-Driven AI Text Detector
https://openreview.net/forum?id=QuDDXJ47nq
Attention Tracker Demo for Prompt Injection Detection
HEx-PHI: Human-Extended Policy-Oriented Harmful Instruction
LLM LabSafety Benchmark
P4D Red-teamer
Resources for ICML 2024 paper "Prompting4Debugging: Red-Teaming Text-to-Image Diffusion Models by Finding Problematic Prompts"
Beyond Demo
Adversarial Example Detector published at ICML 2024: https://proceedings.mlr.press/v235/he24l.html
