PhD Position F/M Theory for Discrete Diffusion Models

inria; France; Híbrido; EUR 20.000 - 27.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Partial reimbursement of public travel

Teleworking possible

Flexible working hours

Social, cultural and sports events

Descripción de la vacante

Inria, the French national research institute for the digital sciences, invites applications for a PhD position under the ERC Starting Grant DYNASTY (Dynamics-Aware Theory of Deep Learning). The project investigates discrete diffusion models, their theoretical foundations, and the impact of neural network-based score estimation.

Travel to conferences may be required, with travel expenses covered. The role emphasizes theory, experiments, and publication of results in top venues.

Formación

• Proven ability to conduct theoretical research in machine learning.
• Experience in conducting experiments for empirical verification.
• Ability to write scientific articles and disseminate work in conferences/journals.
• Strong programming in Python and related DL libraries.

Responsabilidades

• Conduct theoretical research.
• Conduct experiments for empirical verification.
• Write scientific articles and disseminate the scientific work in appropriate venues.

Conocimientos

Python

Deep learning

Descripción del empleo

Inria, the French national research institute for the digital sciences

Organisation/Company Inria, the French national research institute for the digital sciences Research Field Computer science Mathematics Researcher Profile First Stage Researcher (R1) Application Deadline 3 Dec 2026 - 00:00 (UTC) Country France Type of Contract Temporary Job Status Full-time Hours Per Week 38.5 Offer Starting Date 1 Jan 2027 Is the job funded through the EU Research Framework Programme? Horizon 2020 Reference Number 2026-10676 Is the Job related to staff position within a Research Infrastructure? No

Offer Description

The PhD position will be in the framework of the ERC Starting Grant DYNASTY (Dynamics-Aware Theory of Deep Learning).

The position might include traveling to conferences for paper presentation. Travel expenses will be covered within the limits of the scale in force.

Discrete diffusion models have become a leading alternative to autoregressive models for generating text and other categorical data [1, 2]. They generate samples by reversing a Markov chain that progressively corrupts the data, using a score function learned by a neural network [1]. Despite their empirical success, their theoretical understanding remains far less developed than that of their continuous counterparts.

The objective of this PhD is to study the theoretical foundations of discrete diffusion models, and in particular how well they learn and generalize when the score is estimated by a neural network. Building on recent statistical analyses [3, 4] and on generalization bounds for diffusion models [5], the project will investigate how statistical, approximation and optimization errors affect the quality of the generated samples, and will validate the theory through numerical experiments.

References
[1] A. Lou, C. Meng, S. Ermon. Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution. ICML, 2024.
[2] S. S. Sahoo, M. Arriola, Y. Schiff, A. Gokaslan, E. Marroquin, J. T. Chiu, A. Rush, V. Kuleshov. Simple and Effective Masked Diffusion Language Models. NeurIPS, 2024.
[3] Z. Zhang, H. Fu, Z. Yang, M. Wang, T. Zhao, M. Chen. Generalization Bounds for Discrete Diffusion: Statistical Advantage of Masking. ICML, 2026.
[4] L. Blunda, T. Farghly, U. Şimşekli, A. Durmus. Discrete Diffusion Models are Minimax Optimal Under Sparsity. arXiv:2610.19020, 2026.
[5] B. Dupuis, D. Shariatian, M. Haddouche, A. Durmus, U. Şimşekli. Algorithm- and Data-Dependent Generalization Bounds for Diffusion Models. arXiv:2506.03849, 2026.

Conduct theoretical research

Conduct experiments for empirical verification

Write scientific articlesDisseminate the scientific work in appropriate venues.

Languages : High-level of professional/academic English

Coding skills : Good level of coding in Python and related deep learning libraries

Languages FRENCH Level Basic

Languages ENGLISH Level Good

Additional Information

• Partial reimbursement of public transport costs
• Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
• Possibility of teleworking and flexible organization of working hours
• Professional equipment available (videoconferencing, loan of computer equipment, etc.)
• Social, cultural and sports events and activities