Search Results - yunchuan+liu

42 Results Sort By:
Exploiting Vulnerabilities and Security Threats in Retrieval-Augmented Generative Models: The LIAR Attack Framework
Invention Description Retrieval-Augmented Generative (RAG) models boost generative AI’s accuracy by connecting large language models (LLMs) with the most current, external, knowledge sources. RAG models are widely used in fact-checking, information retrieval, and AI-driven search engines. Despite their utility, adversarial threats can exploit...
Published: 1/30/2026   |   Inventor(s): Zhen Tan, Chengshuai Zhao, Raha Moraffah, Huan Liu
Keywords(s): Artificial Intelligence, Data Mining, Large Language Models, Machine Learning, Natural Language Processing
Category(s): Artificial Intelligence/Machine Learning, Cybersecurity, Computing & Information Technology, Intelligence & Security, Physical Science
Transgenic Mouse Model for Marfan Syndrome Research
Description: A genetically modified mouse model carrying the FBN1Q2469X mutation enables advanced study of systemic Marfan Syndrome manifestations and therapeutic development. This innovative transgenic mouse model incorporates a mutation analogous to the human FBN1Q2467X gene mutation linked to Marfan Syndrome, resulting in fibrillin-1 deficiency...
Published: 12/23/2025   |   Inventor(s): Li Li, Shichao Wu, Jiawei Zhao
Keywords(s):  
Category(s): Technology Category > Research tools, Technology Category > Animal models, Technology Category > Biology, Technology Category > Diagnostic
Exploiting Class Probabilities for Black-Box Sentence-Level Attacks
Background Text classification models have become increasingly prevalent in cybersecurity applications, but remain susceptible to adversarial examples (e.g., carefully crafted sentences with human-unrecognizable changes to the inputs, that are misclassified). Adversarial attacks provide profound insights into the classifiers’ vulnerabilities,...
Published: 9/15/2025   |   Inventor(s): Raha Moraffah, Huan Liu
Keywords(s): Machine Learning, Natural Language Processing, Red teaming, Security
Category(s): Physical Science, Applied Technologies, Artificial Intelligence/Machine Learning, Cybersecurity
Adversarial Text Purification: Large Language Model Approach for Defense
Background Adversarial purification is a defense mechanism for safe-guarding classifiers against adversarial attacks without knowing the type of attacks or training of the classifier. These techniques analyze and eliminate adversarial perturbations from the attacked inputs, and help to restore purified samples that retain similarity to the attacked...
Published: 6/27/2025   |   Inventor(s): Raha Moraffah, Shubh Khandelwal, Amrita Bhattacharjee, Huan Liu
Keywords(s): Artificial Intelligence, Defense Applications, Machine Learning, Natural Language Processing, Security, Text Mining
Category(s): Physical Science, Artificial Intelligence/Machine Learning, Applied Technologies, Cybersecurity
Methods and Systems for Detecting Disinformation Generated by Large Language Models
Background The rapid dissemination of news is an important factor for the global population to stay up to date with current events, or to make well-informed decisions in dynamic sectors like the stock market. Journalists typically include large amounts of quantitative information using data visualization to help enhance accessibility and comprehension...
Published: 9/10/2025   |   Inventor(s): Bohan Jiang, Zhen Tan, Ayushi Nirmal, Huan Liu
Keywords(s): Data Mining, Disinformation, Large Language Model, Machine Learning
Category(s): Cybersecurity, Physical Science, Artificial Intelligence/Machine Learning
PMU-based event detection tool for grid analytics and monitoring
Technology Overview: Researchers at the University of Nevada, Reno have developed real-time grid event detection software that analyzes phasor measurement unit (PMU) data. The software uses singular value decomposition (SVD) to measure changes in rank signatures across PMU data signals (voltage, current, frequency). It employs Bayesian optimization...
Published: 5/6/2025   |   Inventor(s): Lei Yang, Amir Ghasemkhani, Yunchuan Liu
Keywords(s):  
Category(s): Technology Classifications > Software, Technology Classifications > Energy & Environment
Enhancing Fairness Through Aleatoric Uncertainty
Background In recent years, advancements in machine learning have revolutionized various domains, but concerns about bias and unfairness in automated decision-making systems have gained significant attention. One of the causes of unfairness in AI systems is the reliance on training data. Aleatoric uncertainty refers to the inherent randomness or variability...
Published: 2/13/2025   |   Inventor(s): Anique Tahir, Lu Cheng, Huan Liu
Keywords(s): Artificial Intelligence, Machine Learning
Category(s): Physical Science, Computing & Information Technology, Intelligence & Security
DNA Methylation Barriers
In normal differentiated cells, most of the genome is densely methylated, except CpG islands near promoters of actively transcribed genes. Maintaining this boundary between unmethylated promoter-associated and adjacent methylated regions is crucial as loss of segregation can lead to disease. Unfortunately, it is not clear where these boundaries are,...
Published: 2/13/2025   |   Inventor(s): Li Liu, Jingmin Shu
Keywords(s):  
Category(s): Applied Technologies, Genomic Assays/Reagents/Tools, Life Science (All LS Techs), Diagnostic Assays/Devices
Novel Anti-HIV Compounds Using Peptides or Peptide Mimetics
Abstract: The subject invention describes a new class of compounds (such as peptides or mimetics) that target viral RNAs and inhibit the viral life cycle by blocking the viral recognition process. More specifically, these compounds are the first against an RNA Target - currently there are no clinical drugs against RNA targets in the treatment of any...
Published: 4/22/2025   |   Inventor(s): Yun-Xing Wang, Liu Yu, Ping Yu, Ina O'Carroll
Keywords(s): Anti-Viral, HIV
Category(s): Collaboration Sought > Licensing, Application > Therapeutics, TherapeuticArea > Infectious Disease
Method and Device for Selectively Labeling RNA
Abstract: Current methods of labeling and synthesizing RNA do not allow for multiple labels or long RNA segments to be synthesized for large RNA on a milligram scale. Investigators at the NCI Structure Biophysics Lab and UT Health Science Center have developed a method to selectively label RNA at specific residues and/or segments using a hybrid solid-liquid...
Published: 4/22/2025   |   Inventor(s): Yun-Xing Wang, Liu Yu, Rui Sousa
Keywords(s): crystallography, Fluorophore labeling, FRET, Isotope labeling, Labeling sensor, NMR Spectroscopy
Category(s): Application > Diagnostics, Collaboration Sought > Licensing, TherapeuticArea > Oncology
1 2 3 4 5