IXAM: Interactive Explainability for Authorship Attribution Models

Milad Alshomary1, Anisha Bhatnagar2, Peter Zeng2, Smaranda Muresan2, Owen Rambow2, Kathleen McKeown1,2,
1Columbia University, New York, USA 2Stony Brook University, New York, USA 2University of Pennsylvania, Philadelphia, USA
arXiv Video Code

IXAM lets you look into the latent space of authorship attribution models

Abstract

We present IXAM, an Interactive eXplainability framework for Authorship Attribution Models. Given an authorship attribution (AA) task and an embedding-based AA model, our tool enables users to interactively explore the model's embedding space and construct an explanation of the model's prediction as a set of writing style features at different levels of granularity. Through a user evaluation, we demonstrate the value of our framework compared to predefined stylistic explanations.

BibTeX

@article{alshomary2025IXAM,
  title={XAM: Interactive Explainability for Authorship Attribution Models},
  author={Alshomary, Milad and Bhatnagar, Anisha and Zeng, Peter and Muresan, Smaranda and Rambow, Owen and McKeown, Kathleen},
  journal={arXiv preprint arXiv:2512.06924},
  year={2025}
}