Meng Jiang
University of Notre Dame, IN·Computer Science·Early-career
Computers can now produce fluent text, but they still struggle to reason like scientists, give expert help in specialized areas, or learn well from limited data. The research aims to make artificial intelligence (AI) systems that reason and assist more reliably.
- Fields
Computer Science
- Topics
natural language processing
natural language generation
abductive reasoning
- Grants
- 3 known grants, through May 2027
What they work on
- Develop and compare natural language generation models, one data-driven and one based on theories of abduction, for comparative summarization and argument generation in scientific domains.
- Improve scientific abductive reasoning with retrieval augmentation from tables, taxonomies and knowledge graphs, and speed it up with knowledge memory networks that store concept representations.
- Build information-extraction methods that turn unstructured social media posts into mental-health ontologies, structured maps of concepts, for efficient retrieval and learning.
- Generate personalized, knowledge-aware support responses for people with anxiety or depression by combining user models, ontologies and contextualized learning from user behavior.
- Augment graph data for graph neural networks using counterfactual inference that treats edges as treatment variables, along with temporal forecasting of graph patterns.
- Improve graph learning through pseudo-labeling and disconnecting nodes in very different communities or clusters, drawing on graph-mining algorithms.
More about the research
Its projects connect knowledge sources, such as knowledge graphs and ontologies, with text generation and graph machine learning. They span scientific writing, mental-health support and better training data for graph models.
Why it matters
These tools could help researchers summarize and compare scientific findings more reliably. They could also give more informed, personalized responses to people with anxiety or depression who seek help on social media. Better graph augmentation could improve predictions on social networks, molecular networks and knowledge graphs, which are used across science and industry.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
III: Small: Intelligent Scientific Text Analytics with Knowledge-Augmented Abductive Reasoning
NSF · Jun 2023 – May 2027
CAREER: Synergistic Approaches for Specialized Intelligent Assistance
NSF · Mar 2022 – Feb 2027
III: Small: Comprehensive Methods to Learn to Augment Graph Data
NSF · Mar 2022 – Feb 2027
Closest labs
Jundong Li
University of Virginia, VA·Computer Science·Early-career
Design graph neural networks tailored to peptides, including amino acid-aware message passing and graph transformers, and combine them with pretrained protein language models to merge structural and evolutionary information.
Jurij Leskovec
Stanford University, CA·Computer Science
Develop multi-level graph neural networks that handle graphs whose nodes are themselves complex structures, such as molecular graphs, for scientific foundation models.
Philip Yu
University of Illinois Chicago, IL·Computer Science
Integrate open-source anomaly detection tools for tabular, graph and time-series data into a single OpenAD ecosystem with a standardized development environment for building and evaluating models.
The school
University of Notre Dame: 19 labs on PI Pages, $346M a year on research, 277 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.