Large Language Models as Anonymizers: An Evaluation on the Text Anonymization Benchmark
收藏Zenodo2026-06-14 更新2026-06-12 收录
下载链接：
https://zenodo.org/doi/10.5281/zenodo.20659360
下载链接
链接失效反馈官方服务：
资源简介：
This Zenodo record provides the LLM anonymization output files supporting the analyses reported in:
Large Language Models as Anonymizers: An Evaluation on the Text Anonymization Benchmark [DOI: to be added upon publication]
Contents: nine CSV files (one per model) containing the original and anonymized texts for all 1,014 TAB documents, along with structural metrics. Input data derived from the TAB dataset (Pilán et al., 2022), publicly available at HuggingFace.
https://huggingface.co/datasets/mattmdjaga/text-anonymization-benchmark-train
License: CC BY 4.0.
For questions or reproducibility issues, contact: alberto.ballesterosr@uah.es
A README included in the record describes the file structure and column definitions.
提供机构：
Zenodo创建时间：
2026-06-12
