Guanhong Tao

Adversarial Robustness in Machine Learning Advanced Malware Detection Techniques Analytical chemistry methods development Anomaly Detection Techniques and Applications Topic Modeling

Publications

140

Citations

3,536

Est. group size

—

Recurring co-author estimate

Active years

Publishing since 1993

Research summary

AI-generated

Guanhong Tao studies the security and robustness of machine learning systems, with a strong focus on identifying and defending against attacks on AI models. Much of the recent work examines 'backdoor attacks' (hidden triggers that make a model misbehave) and safety issues in large language models, vision-language models, and image-generation systems. The research spans both discovering vulnerabilities and building methods to detect or remove them.

AI/machine learning securityBackdoor attacks and defensesRobustness of large language and vision-language modelsAdversarial attacks on learned control policiesContent moderation and safe AI generation

Publication activity grew sharply around 2022 and has remained high, averaging about 14 papers per year over the last five years.

Generated by claude-opus-4-8 from public bibliographic data · Jul 11, 2026

Publication cadence

Publications per year over the last 10 years — averaging 14.0/year recently

Recent publications

Less Is More -- Until It Breaks: Security Pitfalls of Vision Token Compression in Large Vision-Language Models
arXiv (Cornell University) · 2026
Less Is More -- Until It Breaks: Security Pitfalls of Vision Token Compression in Large Vision-Language Models
arXiv (Cornell University) · 2026
PromptGuard: Soft Prompt-Guided Unsafe Content Moderation for Text-to-Image Models
IEEE Transactions on Information Forensics and Security · 2026
Mitigating Backdoor Attacks via Trigger Reconstruction and Model Hardening
2026
Alleviating the Fear of Losing Alignment in LLM Fine-tuning
2025
BAIT: Large Language Model Backdoor Scanning by Inverting Attack Target
2025
Rethinking the Evaluation of Secure Code Generation
arXiv (Cornell University) · 2025
Blackbox Dataset Inference for LLM
arXiv (Cornell University) · 2025
How Vulnerable Is My Learned Policy? Universal Adversarial Perturbation Attacks On Modern Behavior Cloning Policies
arXiv (Cornell University) · 2025
Multi-label learning-based intelligent jamming frequency detection method of frequency-hopping communication
2025
A Systematic Threat Modeling of LLM Applications
2025
Temporal Logic-Based Multi-Vehicle Backdoor Attacks against Offline RL Agents in End-to-end Autonomous Driving
arXiv (Cornell University) · 2025
Profiler: Black-box AI-generated Text Origin Detection via Context-aware Inference Pattern Analysis
2025
Dataset Poisoning Attacks on Behavioral Cloning Policies
arXiv (Cornell University) · 2025
Backdoor Attack on Vision Language Models with Stealthy Semantic Manipulation
arXiv (Cornell University) · 2025

Publishes in

arXiv (Cornell University)×37
2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)×3
Lecture notes in computer science×2
2022 IEEE Symposium on Security and Privacy (SP)×2
Proceedings of the AAAI Conference on Artificial Intelligence×2

This profile was generated automatically from public scholarly data (OpenAlex). Group size and activity levels are estimates derived from co-authorship patterns.

Last updated Jul 11, 2026.

Claim or correct this profile