Assistant Professor
Assistant Professor, School of Computing
Postdoc Scholar, Computational Science and Engineering, Georgia Institute of Technology
Ph.D., Computer Science, University of Texas at Dallas
| Office: | IIT Galvin Tower #15E3-3, 10 West 35th Street, Chicago, IL 60616, USA (Google Map Location) |
| Office Hour: | Thursday 2-4pm |
| Telephone: | Click to show |
| Email: | yhu89 [at] illinoistech [dot] edu |
| Profiles: | Google ScholarLinkedInGitHub |
About
Dr. Yibo Hu is an Assistant Professor in the School of Computing at Illinois Institute of Technology. Before joining IIT, he was a Postdoctoral Scholar in the School of Computational Science and Engineering at the Georgia Institute of Technology.
He works on the safety and reliability of AI systems: when large language models and multi-agent LLM systems can be trusted, and how they fail. His research shows that confidence, agreement, and statistical guarantees stop being reliable safety signals once a model's decisions are shaped by social influence, distribution shift, and multi-agent interaction. He builds methods and benchmarks to measure and contain these failures, with applications in health, social computing, and other high-stakes domains.
News
- Aug 2026: Three new student-led manuscripts are online: Alicia Guerra on measuring conformity in open-ended generation [link], Hanyu Su and Carlota Julbe on subtype robustness [link], and Diego Fernandez Arias on detecting distributed backdoors early [link].
- Aug 2026: Code and data for Safety-Flag, our benchmark for the reliability and calibration of LLM content moderators, are publicly released. [link]
- Jun 2026: Alicia Guerra and Hanyu Su join the group.
- Jun 2026: New student-led manuscripts from Zixian He, on LLM political event coding [link], and Maroof Kousar, on medication dosing under temporal uncertainty [link]. Suraj Babu Thimma Krishnaram also completed work with the group on content moderation.
Research Areas
- Reliability and uncertainty in AI models — when a model’s confidence can be trusted, and what it does when it meets something new
- Safety of multi-agent LLM systems — social influence between models, and harms that appear only across agents
- Social computing — online platforms, social networks, and computational social science
- AI for high-stakes domains — health decision-making and social measurement
Selected Publications
Recent Manuscripts
Social Influence and Collective Decision-Making
Alicia Guerra and Yibo Hu. The Evaluator Is Part of the Experiment: Measuring Open-Ended LLM Conformity. arXiv:2608.04463, 2026.
Yibo Hu and Jiaming Qu. Social Pressure Breaks Majority Voting in LLM Safety Panels. arXiv:2608.04415, 2026. Under review.
Yibo Hu and Jiaming Qu. Most LLM Conformity Needs No Speaker: Measuring the Speaker-Free Floor in Peer-Pressure Benchmarks. arXiv:2607.05545, 2026. Under review.
Jiaming Qu, Lucheng Fu, and Yibo Hu. Easier to Mislead Than to Correct: Harmful and Beneficial Revision in LLM Conformity. arXiv:2606.01637, 2026. Under review.
Monitoring and Security in Multi-Agent Systems
Yibo Hu. Silence Is Endorsement: Verification-Status Laundering in LLM Agent Pipelines. doi:10.5281/zenodo.21907542, 2026. Under review. [pdf]
Diego Fernandez Arias, Dev Prashant Mistry, Ren Wang, and Yibo Hu. Early Detection of Distributed Backdoors in Multi-Agent LLM Systems: A Characterization Study. arXiv:2607.24893, 2026. Under review.
Yibo Hu and Ren Wang. When Local Monitors Miss Compositional Harm: Diagnosing Distributed Backdoors in Multi-Agent Systems. arXiv:2607.11751, 2026. Under review.
Reliability under Uncertainty and Distribution Shift
Yibo Hu. One Axis, No Brake: Self-Knowledge Limits the Filtering of Harmful Peer Conformity in LLMs. doi:10.5281/zenodo.21907262, 2026. Under review. [pdf]
Yibo Hu. Safety-Flag: A Unified Benchmark for the Reliability and Calibration of LLM Content Moderators. doi:10.5281/zenodo.21906739, 2026. Under review. [pdf] [code & data]
Hanyu Su, Carlota Julbe i Juanola, and Yibo Hu. Subtype Robustness Is Not Just Accuracy: Calibration Under Unseen Subtype Shift. arXiv:2608.00928, 2026. Under review.
AI for Health and Public-Interest Domains
Zixian He, Bharath Raahul Murugesan, Patrick T. Brandt, and Yibo Hu. When Better Codebooks Are Not Enough: Predictive Performance and Behavioral Reliability in LLM Political Event Coding. arXiv:2606.06781, 2026. Under review.
Suraj Babu Thimma Krishnaram, Yibo Hu, and Karthikeyan Saravanan. When Surface Form Changes Moderation Decisions: A Paired Study of Code-Mixed Workflow Instability. arXiv:2606.05654, 2026.
Maroof Kousar and Yibo Hu. Can I Take Another Dose? Evaluating LLM Decision-Making Under Temporal Uncertainty in OTC Dosing. arXiv:2606.04262, 2026. Under review.
Peer-Reviewed Publications
Yiqiao Jin, Mohit Chandra, Gaurav Verma, Yibo Hu, Munmun De Choudhury, Srijan Kumar. Better to Ask in English: Cross-Lingual Evaluation of Large Language Models for Healthcare Queries. WWW 2024.
Yibo Hu, Erick Skorupa Parolin, Latifur Khan, Patrick T. Brandt, Javier Osorio, Vito J. D'Orazio. Leveraging Codebook Knowledge with NLI and ChatGPT for Zero-Shot Political Relation Classification. ACL 2024.
Bing He, Yibo Hu, Yeon-Chang Lee, Soyoung Oh, Gaurav Verma, Srijan Kumar. A Survey on the Role of Crowds in Combating Online Misinformation: Annotators, Evaluators, and Creators. ACM TKDD 19(1), 2024.
Yibo Hu, MohammadSaleh Hosseini, Erick Skorupa Parolin, Javier Osorio, Latifur Khan, Patrick Brandt, and Vito D'Orazio. ConfliBERT: A Pre-trained Language Model for Political Conflict and Violence. NAACL 2022.
Yibo Hu, Yu Lin, Erick Skorupa Parolin, Latifur Khan, and Kevin Hamlen. Controllable Fake Document Infilling for Cyber Deception. EMNLP 2022 (Findings).
Yibo Hu, and Latifur Khan. Uncertainty-aware reliable text classification. KDD 2021.
Yibo Hu, Yuzhe Ou, Xujiang Zhao, Jin-Hee Cho, and Feng Chen. Multidimensional Uncertainty-Aware Evidential Neural Networks. AAAI 2021.
Group
I work with students on trustworthy AI and language models, including multi-agent systems, social computing, and high-stakes applications. Please see the Group page for current students and alumni. If you are interested in joining, please see the Prospective Students page.
Teaching
- ITMD 526 / CS 520: Data Warehousing (Graduate), Illinois Tech, Fall 2026
- ITMD 524: Applied Artificial Intelligence and Deep Learning (Graduate), Illinois Tech, Fall 2025, Spring 2026
- ITMD 523: Advanced Topics in Data Management (Graduate), Illinois Tech, Spring 2026
- CSE 8803 DSN: Data Science for Social Networks (Graduate), Georgia Tech, Fall 2023
Service
Editorial Roles
- Co-Editor, Graph-Based Retrieval-Augmented Generation Systems, Frontiers in Big Data (2025-2026) [link]
Program Committee & Reviewing
- ACL (2023-), EMNLP (2022-), NAACL (2024-), WWW (2024-)
- IJCAI (Distinguished PC Member, 2023), KDD (2020-)
- SDM (2022-), DASFAA (2023), IJCNN (2024), DSAA (2025)
- Data4SoftSec 2026 (Workshop on Datasets for Software Security @ IEEE S&P)