Marijan Soric

Full name: Marijan Sorić

I am a Ph.D. candidate at Inria Valda (Inria Paris, DI ENS, CNRS), Inria CEDAR (Inria Saclay, LIX, CNRS) and École normal supérieure (ENS - PSL University).
The Ph.D. is funded by the GéolAug project, coming from a collaboration between Inria and BRGM (French National Geological Survey).
PhD: Exploitation and Structuring of Heterogeneous Geological Data and Knowledge.
Thesis supervisors: Pierre Senellart, Ioana Manolescu and Cécile Gracianne.

Previously at École Centrale de Lyon (2021–2024) and NTNU (2024, exchange).


Research

Conference Papers

  1. Benchmarking Table Extraction from Heterogeneous Scientific PDF Documents.
    In: Proceedings of the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2026). ACM (2026). [Paper][Code][Dataset][Website]
    Marijan Soric, Cécile Gracianne, Ioana Manolescu, Pierre Senellart.

Preprints

  1. Benchmarking Table Extraction from Heterogeneous Scientific Documents.
    arXiv (2025). [Preprint][Code][Dataset][Website]
    Marijan Soric, Cécile Gracianne, Ioana Manolescu, Pierre Senellart.
  2. Understanding and Extracting Table Information from BRGM Documents.
    Master Thesis Report, École Centrale de Lyon / Inria (2025). [HAL][Report][Slides]
    Marijan Soric

Talks

  1. Benchmark: Extracting Table Information from Scientific Documents.
    Valda Seminar, ENS, Paris, 23 May 2025. [Slides][Link]

Posters

  1. Towards Probabilistic Georeferencing of Geological Maps from PDF Reports.
    KDD 2026 PhD Consortium, 2026. [Paper]
  2. Benchmarking Table Extraction from Heterogeneous Scientific Documents.
    BRGM GéoNum Seminar, 2025. [Poster]

Teaching


Professional/Academic Service


Contact

My email: firstname [dot] lastname [at] inria [dot] fr
Office:
DI - Ecole Normale Supérieure
45, rue d'Ulm
75 230 Paris Cedex 05 - France
Office: Escalier A, Hauts du DI, Bureau 4

Last modified: (Paris)