Martin Sedláček

CIIRC CTU, B-642

Jugoslávských partyzánů 1580/3, 160 00

Prague, Czech Republic

I'm a first year student in the ELLIS PhD programme co-advised by Josef Sivic (CIIRC) and Cordelia Schmid (INRIA Paris) working on robot learning and computer vision. I currently focus on generalization capabilities of VLAs - see our recent benchmarking work - REALM.

Previously, I completed the MSc in AI at the University of Amsterdam awarded with the "Cum Laude" distinction and received the ELLIS Honours Research scholarship.

Between 2018 and 2023, I also worked as a Software Engineer (Palantir, Amazon, Goldman Sachs) and simultaneously completed my undergraduate studies in Computer Science and Mathematics. I graduated with the EECS Final Year Prize (top 5% in the department) from Queen Mary University of London (QMUL) and received full funding for my studies as part of a degree apprenticeship with Goldman Sachs.

News

01/09/2025	Started as an ELLIS PhD at CIIRC in Prague working with Josef, Vladimir, and Cordelia.
15/06/2025	Graduated Cum Laude with an MSc. in AI from UvA.
01/09/2024	Accepted to the ELLIS Honours Research Fellowship co-advised by Josef Sivic and Cees Snoek focusing on generalization in VLAs.
17/07/2024	Blogpost on consistency models for molecular structure prediction accepted to the GRaM Workshop at ICML 2024.
28/05/2024	Reproducibility study accepted to MLRC'23@NeurIPS and jointly published in TMLR.

Selected publications

REALM: A Real-to-Sim Validated Benchmark for Generalization in Robotic Manipulation

Martin Sedlacek*, Pavlo Yefanov*, Georgy Ponimatkin, Jai Bardhan, Simon Pilc, Mederic Fourmy, Evangelos Kazakos, Cees G. M. Snoek, Josef Sivic, Vladimir Petrik

arXiv, 2025

Bib PDF GitHub

@article{sedlacek2025realm,
    title = {REALM: A Real-to-Sim Validated Benchmark for Generalization in Robotic Manipulation},
    show_bib = {true},
    author = {Martin Sedlacek, Pavlo Yefanov, Georgy Ponimatkin, Jai Bardhan, Simon Pilc, Mederic Fourmy, Evangelos Kazakos, Cees G. M. Snoek, Josef Sivic, Vladimir Petrik},
    year = {2025},
    journal = {arXiv preprint arXiv:2512.19562}
}

Equivariant Diffusion for Molecule Generation in 3D using Consistency Models

Martin Sedlacek* Antonis Vozikis*, Patrik Bartak, Luke Cadigan, and Mohan Guo

GRaM Workshop @ ICML (Blogpost Track), 2024

Bib Website

@article{sedlacek2024edmConsistency,
    title = {Equivariant Diffusion for Molecule Generation in 3D using Consistency Models},
    show_bib = {true},
    author = {Sedlacek, Martin and Vozikis, Antonis and Bartak, Patrik and Cadigan, Luke and Guo, Mohan},
    year = {2024},
    journal = {GRaM Workshop, ICML}
}

Reproducibility study of "LICO: Explainable Models with Language-Image Consistency"

Luan Fletcher*, Robert van der Klis*, Martin Sedlacek*, Stefan Vasilev*, and Christos Athanasiadis

TMLR, 2024

Bib PDF GitHub

@article{fletcher2024lico_repro,
    title = {Reproducibility study of "LICO: Explainable Models with Language-Image Consistency"},
    show_bib = {true},
    author = {Fletcher, Luan and Klis, Robert and Sedlacek, Martin and Vasilev, Stefan and Athanasiadis, Christos},
    year = {2024},
    journal = {TMLR}
}