@article{whenllmsbenchmarkthemselvesdeconstructin,
  title = {When LLMs Benchmark Themselves: Deconstructing Self-Bias in Automated Evaluation},
  author = {Wenda Xu and Sweta Agrawal and Vilém Zouhar and Markus Freitag and Daniel Deutsch},
  year = {2025},
  eprint = {2509.26600},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2509.26600},
}