Mathilde Perez

mathilde.perez@telecom-paris.fr

Télécom Paris

19 place Marguerite Perey

91120 Palaiseau, France

I am a 3rd year PhD student at Télécom Paris, member of the S²A team (Signal, Statistics and Machine Learning), where I work at the intersection of graph learning and fairness in AI. My research explores the short and long-term implications of fairness constraints on evolutionary graph learning models, and aims to design an evaluation framework to compare graph structures resulting from biased and unbiased link prediction models.

Before my PhD, I worked as a Research Engineer at Télécom Paris on temporal graph simulation with Hawkes processes. I also completed internships at ISIR (Sorbonne/CNRS) on social attention modeling in conversational agents, and at CerCo (CNRS) on the similarities between CLIP and the human brain.

I hold a Master’s degree in Applied Mathematics (MApi3) from Université Paul Sabatier, Toulouse.

news

Jul 19, 2026	🌟 The same paper, How Predicted Links Influence Network Evolution: Disentangling Choice and Algorithmic Feedback in Dynamic Graphs, has also been selected for a spotlight presentation at ECAF 2026!
Jul 06, 2026	📢 I have a paper accepted at CAp 2026 at Université de Montpellier, France.
Jun 15, 2026	🎤 I will be speaking at the EUNIC Science Day in Madrid on how social media algorithms reinforce inequality.
Jun 01, 2026	🎉 My paper How Predicted Links Influence Network Evolution: Disentangling Choice and Algorithmic Feedback in Dynamic Graphs has been accepted at UAI 2026!

selected publications

DAMI

SimHawNet: a modified Hawkes process for temporal network simulation

Mathilde Perez, Raphaël Romero, Bo Kang, and 3 more authors

Data Mining and Knowledge Discovery, 2025

DOI Bib HTML

@article{perez2025simhawnet,
  title = {SimHawNet: a modified Hawkes process for temporal network simulation},
  author = {Perez, Mathilde and Romero, Rapha{\"e}l and Kang, Bo and De Bie, Tijl and Lijffijt, Jefrey and Laclau, Charlotte},
  journal = {Data Mining and Knowledge Discovery},
  volume = {39},
  number = {5},
  year = {2025},
  doi = {10.1007/s10618-025-01119-1},
  url = {https://doi.org/10.1007/s10618-025-01119-1},
}

arXiv

TopoFair: Linking Topological Bias to Fairness in Link Prediction Benchmarks

Lilian Marey, Mathilde Perez, Tiphaine Viard, and 1 more author

2026

Preprint

Bib HTML

@misc{marey2026topofair,
  title = {TopoFair: Linking Topological Bias to Fairness in Link Prediction Benchmarks},
  author = {Marey, Lilian and Perez, Mathilde and Viard, Tiphaine and Laclau, Charlotte},
  year = {2026},
  archiveprefix = {arXiv},
  primaryclass = {cs.LG},
  url = {https://arxiv.org/abs/2602.11802},
  note = {Preprint},
}

arXiv

How Predicted Links Influence Network Evolution: Disentangling Choice and Algorithmic Feedback in Dynamic Graphs

Mathilde Perez, Raphaël Romero, Jefrey Lijffijt, and 1 more author

2026

Accepted at UAI 2026

Bib HTML

@misc{perez2026predicted,
  title = {How Predicted Links Influence Network Evolution: Disentangling Choice and Algorithmic Feedback in Dynamic Graphs},
  author = {Perez, Mathilde and Romero, Rapha{\"e}l and Lijffijt, Jefrey and Laclau, Charlotte},
  year = {2026},
  archiveprefix = {arXiv},
  primaryclass = {cs.LG},
  url = {https://arxiv.org/abs/2603.03945},
  note = {Accepted at UAI 2026},
}