Arijit Ghosh

I am a (happy) 2nd year Ph.D. student in Computer Vision at Institut Polytechnique de Paris (IMAGINE Lab) under the supervision of David Picard. My main research interest lies in multi-modal generative models and the generalization of such multimodal models. Prior to my Ph.D. I completed my M.Sc. in Artificial Intelligence from University Erlangen-Nürnberg, Germany.

Publications

MIRO teaser

MIRO: MultIReward cOnditioned pretraining improves T2I quality and efficiency

Nicolas Dufour, Lucas Degeorge, Arijit Ghosh, Vicky Kalogeiton, David Picard

* Equal contribution

ICML, 2026

PoM teaser

PoM: A Linear-Time Replacement for Attention with the Polynomial Mixer

David Picard, Nicolas Dufour, Lucas Degeorge, Arijit Ghosh, Davide Allegro, Tom Ravaud, Yohann Perron, Corentin Sautier, Zeynep Sonat Baltaci, Fei Meng, Syrine Kalleli, Marta López-Rauhut, Thibaut Loiseau, Ségolène Albouy, Raphael Baena, Elliot Vincent, Loic Landrieu

CVPR Findings, 2026

How Far Can We Go teaser

How far can we go with ImageNet for Text-to-Image generation?

Lucas Degeorge, Arijit Ghosh, Nicolas Dufour, David Picard, Vicky Kalogeiton

* Equal contribution

ArXiV, 2025

Education

Ph.D. in Computer Vision

Institut Polytechnique de Paris

2024 – 2027

M.Sc. in Artificial Intelligence

University Erlangen-Nürnberg

Grade: 1.30 (Best: 1.0, Worst: 4.0)

2021 – 2024

B.Tech. in Electronics and Communication Engineering

Maulana Abul Kalam Azad National Institute of Technology, West Bengal

Grade: 9.24 (Best: 10.0, Worst: 5.0)

2017 – 2021

Experience

Research Assistant

IMAGINE Lab, IP Paris

Jun 2024 – Nov 2024

Student Research Assistant

IDEA Lab, University Erlangen-Nürnberg

Jan 2022 – May 2024

Student Research Assistant

Fraunhofer IIS, Nürnberg

Mar 2023 – May 2024

Talks

Flow Matching Tutorial

Flow Matching Tutorial

Weakly Supervised Learning

Weakly Supervised Learning Talk

Representation Learning

Representation Learning Talk

Teaching

Teaching Assistant

University Erlangen-Nürnberg

Algorithms, Programming and Data Representation