Publications
Google Scholar · DBLP · Download BibTeX
2026
Yanyi Pu, Damian A. Gonzalez-Salzberg, Zheng Yuan, and Nikolaos Aletras
EMNLP 2026 · Main Conference · Accepted
A benchmark for predicting non-pecuniary damage awards from European Court of Human Rights case information, comparing feature-based, retrieval, language-model, and agentic approaches.
BibTeX
@misc{pu2026ecthrnpd,
title = {How Much is a Human Right Worth? {ECtHR-NPD}: A Benchmark for Predicting Non-Pecuniary Damage Awards},
author = {Pu, Yanyi and Gonzalez-Salzberg, Damian A. and Yuan, Zheng and Aletras, Nikolaos},
year = {2026},
eprint = {2609.18908},
archivePrefix = {arXiv},
primaryClass = {cs.CL},
url = {https://arxiv.org/abs/2609.18908},
note = {Accepted to the EMNLP 2026 Main Conference. Citation points to the available arXiv version.}
}
2023
Xiangjian Jiang* and Yanyi Pu*
* Equal contribution.
Temporal Graph Learning Workshop at NeurIPS 2023
An empirical study of how time granularity, model architecture, and negative sampling affect dynamic link prediction.
BibTeX
@misc{jiang2023timegranularity,
title = {Exploring Time Granularity on Temporal Graphs for Dynamic Link Prediction in Real-world Networks},
author = {Jiang, Xiangjian and Pu, Yanyi},
year = {2023},
eprint = {2311.12255},
archivePrefix = {arXiv},
primaryClass = {cs.LG},
url = {https://arxiv.org/abs/2311.12255},
note = {Presented at the Temporal Graph Learning Workshop at NeurIPS 2023. Xiangjian Jiang and Yanyi Pu contributed equally.}
}
2022
Amyra Meidiana, Seok-Hee Hong, Yanyi Pu, Justin Lee, Peter Eades, and Jinwook Seo
EuroVis 2022
Spectral sparsification methods and evaluation measures for visualising large dynamic graphs while retaining important structural changes.
BibTeX
@inproceedings{meidiana2022dss,
title = {{DSS}: Drawing Dynamic Graphs with Spectral Sparsification},
author = {Meidiana, Amyra and Hong, Seok-Hee and Pu, Yanyi and Lee, Justin and Eades, Peter and Seo, Jinwook},
booktitle = {EuroVis 2022 - Short Papers},
year = {2022},
pages = {55--59},
publisher = {The Eurographics Association},
doi = {10.2312/evs.20221093},
url = {https://doi.org/10.2312/evs.20221093}
}