# Yizirui Fang

> Easey (Yizirui) Fang is an Amazon SDE and applied ML researcher focused on LLM agents, embodied AI, uncertainty-aware decision making, and rigorous model evaluation.

This is Yizirui (Easey) Fang's personal research and engineering website.
Publication records include abstracts, citations, and links to papers and code.
The TMLR 2026 and NeurIPS 2024 workshop entries for Learning to Defer are versions of the same research.

## Start here

- [Research topics](https://yiziruifang.com/research/): find papers by research question, method, and scope.
- [Research topics in Markdown](https://yiziruifang.com/research/index.md)

- [Paper catalog with complete abstracts](https://yiziruifang.com/papers.json): JSON records containing authors, venue, abstract source, PDF, and BibTeX URLs.
- [Profile](https://yiziruifang.com/): biography, research interests, selected work, and profile links.
- [Profile in Markdown](https://yiziruifang.com/index.md): a text version generated from the same content.

- [Projects](https://yiziruifang.com/projects/): Machine learning projects by Yizirui Fang: learning to defer, embodied instruction following, human-centered AI, and conformal prediction, with papers and public artifacts.
- [Projects in Markdown](https://yiziruifang.com/projects/index.md)

- [Experience](https://yiziruifang.com/experience/): Yizirui Fang’s engineering and research experience in LLM agents, embodied AI, uncertainty quantification, and human-AI collaboration.
- [Experience in Markdown](https://yiziruifang.com/experience/index.md)


## Projects

- [Learning to Defer with an Uncertain Rejector via Conformal Prediction](https://yiziruifang.com/project/uncertain-defer/): TMLR 2026 paper on uncertainty-aware learning to defer, using conformal prediction to make human-AI routing safer under rejector uncertainty and distribution shift.
- [Project in Markdown](https://yiziruifang.com/project/uncertain-defer/index.md)

- [Pragmatic Embodied Spoken Instruction Following in Human-Robot Collaboration with Theory of Mind](https://yiziruifang.com/project/siftom/): ICRA 2026 paper on spoken instruction following for embodied agents using theory-of-mind inference over speech, human perception, and robot goals.
- [Project in Markdown](https://yiziruifang.com/project/siftom/index.md)

- [ARCHED: Human-Centered AI-Assisted Instructional Design](https://yiziruifang.com/project/arched/): A human-centered framework for transparent, responsible, and collaborative AI-assisted instructional design.
- [Project in Markdown](https://yiziruifang.com/project/arched/index.md)

- [Investigating Data Usage for Inductive Conformal Predictors](https://yiziruifang.com/project/conformal-predictors/): Studying how data allocation choices affect inductive conformal prediction, calibration behavior, and uncertainty guarantees.
- [Project in Markdown](https://yiziruifang.com/project/conformal-predictors/index.md)


## Publications

- [Learning to Defer with an Uncertain Rejector via Conformal Prediction — 2026](https://yiziruifang.com/publication/fang-2026-learning-tmlr/): Transactions on Machine Learning Research (TMLR). Conformal prediction quantifies uncertainty in learning-to-defer rejectors, enabling abstention and human-model consensus workflows for image and hate-speech classification.
- [Publication in Markdown](https://yiziruifang.com/publication/fang-2026-learning-tmlr/index.md)

- [Pragmatic Embodied Spoken Instruction Following in Human-Robot Collaboration with Theory of Mind — 2026](https://yiziruifang.com/publication/ying-2024-siftom/): IEEE International Conference on Robotics and Automation (ICRA 2026). SIFToM combines vision-language models with probabilistic theory-of-mind inference for noisy spoken instruction following in human-robot collaboration.
- [Publication in Markdown](https://yiziruifang.com/publication/ying-2024-siftom/index.md)

- [ARCHED: A Human-Centered Framework for Transparent, Responsible, and Collaborative AI-Assisted Instructional Design — 2025](https://yiziruifang.com/publication/li-2024-arched/): Proceedings of the Innovation and Responsibility in AI-Supported Education Workshop (iRAISE 2025), PMLR 273:94–104. Human-centered AI workflow for learning-objective generation, Bloom's taxonomy analysis, and educator-guided instructional design.
- [Publication in Markdown](https://yiziruifang.com/publication/li-2024-arched/index.md)

- [Learning to Defer with an Uncertain Rejector via Conformal Prediction — 2024](https://yiziruifang.com/publication/fang-2024-learning-workshop/): NeurIPS 2024 Workshop on Bayesian Decision-making and Uncertainty. NeurIPS 2024 workshop paper on conformal uncertainty sets for learning-to-defer rejectors, with abstention and human-model consensus on object and hate-speech detection tasks.
- [Publication in Markdown](https://yiziruifang.com/publication/fang-2024-learning-workshop/index.md)

- [Investigating Data Usage for Inductive Conformal Predictors — 2024](https://yiziruifang.com/publication/fang-2024-investigating/): arXiv preprint arXiv:2406.12262. How training/calibration splits and overlap affect coverage validity and prediction-set size in inductive conformal classification.
- [Publication in Markdown](https://yiziruifang.com/publication/fang-2024-investigating/index.md)


## Additional resources

- [All publication records](https://yiziruifang.com/publication/)
- [XML sitemap](https://yiziruifang.com/sitemap.xml)
- [RSS feed](https://yiziruifang.com/index.xml)
- [Website source](https://github.com/yizirui/yizirui.github.io)