Qiaozhu Mei
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Qiaozhu Mei has a total of 190 co-authors. Here are the top co-authors by number of shared publications:
- Jiaqi Ma
School of Information University of Michigan Ann Arbor Michigan USA
8 shared publications - Yutong Xie
University of Michigan, Ann Arbor, MI, USA
7 shared publications - David A. Hanauer
Michigan Institute for Clinical and Health Research, University of Michigan Medical School , Ann Arbor, MI, 48109
5 shared publications - Jiaqi Ma
UCLA
5 shared publications - Matthew O. Jackson
Department of Economics, Stanford University
5 shared publications Cheng Li
Google research
4 shared publications
Latest Publications
- Open Access
- 22 August 2026
- Published by Springer Nature in Humanities and Social Sciences Communications
Using emoji reduces dropout of remote workers: a causal analysis on GitHub
- 22 August 2026
- Published by Springer Nature in Humanities and Social Sciences Communications
Open AccessUsing emoji reduces dropout of remote workers: a causal analysis on GitHub
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 12492-12503
Distributional Alignment for Social Simulation with LLMs: A Mixture Modeling Approach
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 12492-12503
Open AccessDistributional Alignment for Social Simulation with LLMs: A Mixture Modeling Approach
- Open Access
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 5822-5833
Pathways of Thoughts: Multi-Directional Thinking for Long-form Personalized Question Answering
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 5822-5833
Open AccessPathways of Thoughts: Multi-Directional Thinking for Long-form Personalized Question Answering
- Open Access
FlyAOC: Evaluating Agentic Ontology Curation of Drosophila Scientific Knowledge Bases
Scientific knowledge bases accelerate discovery by curating findings from primary literature into structured, queryable formats for both human researchers and emerging AI systems. Maintaining these resources requires expert curators to search relevant papers, reconcile evidence across documents, and produce ontology-grounded annotations - a workflow that existing benchmarks, focused on isolated subtasks like named entity recognition or relation extraction, do not capture. We present FlyBench to evaluate AI agents on end-to-end agentic ontology curation from scientific literature. Given only a gene symbol, agents must search and read from a corpus of 16,898 full-text papers to produce structured annotations: Gene Ontology terms describing function, expression patterns, and historical synonyms linking decades of nomenclature. The benchmark includes 7,397 expert-curated annotations across 100 genes drawn from FlyBase, the Drosophila (fruit fly) knowledge base. We evaluate four baseline agent architectures: memorization, fixed pipeline, single-agent, and multi-agent. We find that architectural choices significantly impact performance, with multi-agent designs outperforming simpler alternatives, yet scaling backbone models yields diminishing returns. All baselines leave substantial room for improvement. Our analysis surfaces several findings to guide future development; for example, agents primarily use retrieval to confirm parametric knowledge rather than discover new information. We hope FlyBench will drive progress on retrieval-augmented scientific reasoning, a capability with broad applications across scientific domains.
Open AccessFlyAOC: Evaluating Agentic Ontology Curation of Drosophila Scientific Knowledge Bases
Scientific knowledge bases accelerate discovery by curating findings from primary literature into structured, queryable formats for both human researchers and emerging AI systems. Maintaining these resources requires expert curators to search relevant papers, reconcile evidence across documents, and produce ontology-grounded annotations - a workflow that existing benchmarks, focused on isolated subtasks like named entity recognition or relation extraction, do not capture. We present FlyBench to evaluate AI agents on end-to-end agentic ontology curation from scientific literature. Given only a gene symbol, agents must search and read from a corpus of 16,898 full-text papers to produce structured annotations: Gene Ontology terms describing function, expression patterns, and historical synonyms linking decades of nomenclature. The benchmark includes 7,397 expert-curated annotations across 100 genes drawn from FlyBase, the Drosophila (fruit fly) knowledge base. We evaluate four baseline agent architectures: memorization, fixed pipeline, single-agent, and multi-agent. We find that architectural choices significantly impact performance, with multi-agent designs outperforming simpler alternatives, yet scaling backbone models yields diminishing returns. All baselines leave substantial room for improvement. Our analysis surfaces several findings to guide future development; for example, agents primarily use retrieval to confirm parametric knowledge rather than discover new information. We hope FlyBench will drive progress on retrieval-augmented scientific reasoning, a capability with broad applications across scientific domains.
- 18 November 2025
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 172, 104952
SemNovel – A new approach to detecting semantic novelty of biomedical publications using embeddings of large language models
- 18 November 2025
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 172, 104952
SemNovel – A new approach to detecting semantic novelty of biomedical publications using embeddings of large language models
- Open Access
- 28 August 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122  (35)
Using large language models to categorize strategic situations and decipher motivations behind human behaviors
By varying prompts to a large language model, we can elicit the full range of human behaviors in a variety of different scenarios in classic economic games. By analyzing which prompts elicit which behaviors, we can categorize and compare different strategic situations, which can also help provide insight into what different economic scenarios might induce people to think about. We discuss how this provides a step toward a nonstandard method of inferring (deciphering) the motivations behind the human behaviors. We also show how this deciphering process can be used to categorize differences in the behavioral tendencies of different populations.
- 28 August 2025
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 122  (35)
Open AccessUsing large language models to categorize strategic situations and decipher motivations behind human behaviors
By varying prompts to a large language model, we can elicit the full range of human behaviors in a variety of different scenarios in classic economic games. By analyzing which prompts elicit which behaviors, we can categorize and compare different strategic situations, which can also help provide insight into what different economic scenarios might induce people to think about. We discuss how this provides a step toward a nonstandard method of inferring (deciphering) the motivations behind the human behaviors. We also show how this deciphering process can be used to categorize differences in the behavioral tendencies of different populations.
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 3843-3854
MapExplorer : New Content Generation from Low-Dimensional Visualizations
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 3843-3854
Open AccessMapExplorer : New Content Generation from Low-Dimensional Visualizations
- 20 June 2025
- Published by Springer Nature
- p. 195-209
Leveraging Group-Level Signals for Robust Many-Domain Generalization
- 20 June 2025
- Published by Springer Nature
- p. 195-209
Leveraging Group-Level Signals for Robust Many-Domain Generalization
- Open Access
Be.FM: Open Foundation Models for Human Behavior
Despite their success in numerous fields, the potential of foundation models for modeling and understanding human behavior remains largely unexplored. We introduce Be.FM, one of the first open foundation models designed for human behavior modeling. Built upon open-source large language models and fine-tuned on a diverse range of behavioral data, Be.FM can be used to understand and predict human decision-making. We construct a comprehensive set of benchmark tasks for testing the capabilities of behavioral foundation models. Our results demonstrate that Be.FM can predict behaviors, infer characteristics of individuals and populations, generate insights about contexts, and apply behavioral science knowledge.
Open AccessBe.FM: Open Foundation Models for Human Behavior
Despite their success in numerous fields, the potential of foundation models for modeling and understanding human behavior remains largely unexplored. We introduce Be.FM, one of the first open foundation models designed for human behavior modeling. Built upon open-source large language models and fine-tuned on a diverse range of behavioral data, Be.FM can be used to understand and predict human decision-making. We construct a comprehensive set of benchmark tasks for testing the capabilities of behavioral foundation models. Our results demonstrate that Be.FM can predict behaviors, infer characteristics of individuals and populations, generate insights about contexts, and apply behavioral science knowledge.
- Open Access
ExAnte: A Benchmark for Ex-Ante Inference in Large Language Models
Large language models (LLMs) face significant challenges in ex-ante reasoning, where analysis, inference, or predictions must be made without access to information from future events. Even with explicit prompts enforcing temporal cutoffs, LLMs often generate outputs influenced by internalized knowledge of events beyond the specified cutoff. This paper introduces a novel task and benchmark designed to evaluate the ability of LLMs to reason while adhering to such temporal constraints. The benchmark includes a variety of tasks: stock prediction, Wikipedia event prediction, scientific publication prediction, and Question Answering (QA), designed to assess factual knowledge under temporal cutoff constraints. We use leakage rate to quantify models' reliance on future information beyond cutoff timestamps. Experimental results reveal that LLMs struggle to consistently adhere to temporal cutoffs across common prompting strategies and tasks, demonstrating persistent challenges in ex-ante reasoning. This benchmark provides a potential evaluation framework to advance the development of LLMs' temporal reasoning ability for time-sensitive applications.
Open AccessExAnte: A Benchmark for Ex-Ante Inference in Large Language Models
Large language models (LLMs) face significant challenges in ex-ante reasoning, where analysis, inference, or predictions must be made without access to information from future events. Even with explicit prompts enforcing temporal cutoffs, LLMs often generate outputs influenced by internalized knowledge of events beyond the specified cutoff. This paper introduces a novel task and benchmark designed to evaluate the ability of LLMs to reason while adhering to such temporal constraints. The benchmark includes a variety of tasks: stock prediction, Wikipedia event prediction, scientific publication prediction, and Question Answering (QA), designed to assess factual knowledge under temporal cutoff constraints. We use leakage rate to quantify models' reliance on future information beyond cutoff timestamps. Experimental results reveal that LLMs struggle to consistently adhere to temporal cutoffs across common prompting strategies and tasks, demonstrating persistent challenges in ex-ante reasoning. This benchmark provides a potential evaluation framework to advance the development of LLMs' temporal reasoning ability for time-sensitive applications.