AI & ML interests
Research Demos and Tools for Trustworthy and Safe AI Development and Deployment
Recent Activity
View all activity
Organization Card
- Welcome to TrustSafeAI! We are a reseach group focusing on evaluating and improving AI safety.
- If you are interested in joining us, please reach out to Pin-Yu Chen
- Team Members and Projects:
| Member | Project | Webpage |
|---|---|---|
| Xiaomeng Hu | RADAR (NeurIPS'23), Gradient Cuff (NeurIPS'24), Token Hilighter (AAAI'25) | webpage |
| Lei Hsiung | NeuralFuse (NeurIPS'24), NCTV (TMLR; AAAI'23), CARBEN (CVPR'23; IJCAI'22) | webpage |
| Zhi-Yi Chin | P4D (ICML'24) | webpage |
| Barry Xiong | DPP (ACL'25) | - |
| Zaitang Li | GREAT Score (NeurIPS'24), Retention Score (AAAI'25) | - |
| Yung-Chen Tang | NCTV (TMLR; AAAI'23) , LLM-Physical-Safety (Communications of the ACM 2026) | webpage |
| Zhiyuan He | BEYOND (ICML'24), RIGID (TMLR 2026) | - |
| Yujun Zhou | LLM LabSafety (Nature Machine Intelligence 2026) | webpage |
| Xiangyu Qi | LLM Finetuning Safety (ICLR'24) | webpage |
| Kuo-Han (Johnson) Hung | Attention Tracker (NAACL'25) | webpage |
| Xiang Li | AudioPerturber, Audio-Deepfake-Detector (ACM CODASPY'26) | webpage |
| Advik Raj Basani | DivEye AI Text Detector (TMLR 2026) | webpage |
| Gert Lek | DetectiveSAM (ICLR'26) | webpage |
| Pin-Yu Chen | All (research supervisor) | webpage |
spaces 15
Running
3
NCTV: Neural Clamping Toolkit and Visualization
🦀
Model-agnostic Toolkit for Neural Network Calibration
Running
Test Time Calibration
💻
Test-time calibration for improving test-time reasoning
Running
LLM Physical Safety
👀
LLM benchmark for Physical Safety
Running
CoP Agentic Red-teaming
🧠
Generate jailbreak prompts for LLMs using principles
Sleeping
Agents
AudioDeepfakeDetector
👀
Detect fake audio using uploaded files