About
I’m Fan Huang, a Ph.D. candidate in Informatics at Indiana University Bloomington. I study human–AI alignment: probing and mitigating the biases and limitations of large language models. Here I share my ongoing work, with articles that explain what they find in plain language.
Articles and papers
AI Explain Tech Blog has two kinds of posts. Articles explain research in plain language. Papers are the research itself, published as preprints with a PDF, a citation and a full version history.
Papers have not been peer reviewed. A paper published here may later appear, revised, in a journal or at a conference.
How papers are published
- Each paper gets a permanent number, such as AIE-WP-2026-001, and a page whose address never changes.
- Revisions are published as new versions. Every earlier version stays online, unchanged.
- Each PDF is fingerprinted with SHA-256 when it is added, and the site refuses to publish a file that no longer matches its fingerprint.
- When a version is first published, a GitHub release records the PDF and its fingerprint, and the release cannot be changed afterwards. All releases are listed on GitHub.
To check a download, compute its SHA-256 fingerprint (shasum -a 256 paper.pdf on macOS and Linux, Get-FileHash paper.pdf -Algorithm SHA256 in Windows PowerShell) and compare it with the value on the paper’s page.
Citing a paper
Each paper’s page has a formatted citation and a BibTeX record. Please cite the version you read, since versions can differ.
Licences
Each paper states its licence on its page. Unless a paper says otherwise, it is published under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence.