Clara Lacroce

Profile picture 

Postdoctoral researcher,
Reasoning and Learning Lab, Mila,
School of Computer Science, McGill University,


*corresponding author.

  • Lacroce C.*, Balle B., Panangaden P., Rabusseau G. “Optimal Approximate Minimization of One-Letter Weighted Finite Automata”, under review in Mathematical Structures in Computer Science

  • Michael Rizvi*, Maude Lizaire, Lacroce C., Rabusseau G. “Simulating weighted automata over sequences and trees with transformers”, to appear at AISTATS 2024

  • Volodimir Mitarchuck*, Lacroce C., Remi Emonet, Remi Eyraud, Amaury Habrard, Rabusseau G. “Length independent PAC-Bayes bound for Simple RNNs”, to appear at AISTATS 2024

  • Lacroce C.* “The approximate minimization problem of weighted finite automata and applications to language modelling: an approach based on Adamyan-Arov-Krein theory”, (PhD Thesis)
    [thesis] [pdf]

  • Lacroce C.*, Panangaden P., Rabusseau G. “Towards an AAK Theory Approach to Approximate Minimization in the Multi-Letter Case”, LEARNAUT 2022

  • Lacroce C.*, Panangaden P., Rabusseau G. “Extracting Weighted Automata for Approximate Minimization in Language Modelling”, ICGI 2020-2021
    [proceedings] [arXiv] [pdf]

  • Balle B., Lacroce C.*, Panangaden P., Precup D., Rabusseau G. “Optimal Spectral-Norm Approximate Minimization of Weighted Finite Automata”, ICALP 2021, 48th International Colloquium on Automata, Languages, and Programming
    [proceedings] [arXiv] [pdf]

  • Lacroce C.* “Deformations of Galois Representations”, (Master Thesis)

  • Papers available here may be subject to copyright and are intended for personal, non-commercial use only.