Internship Reinforcement learning of VLMs for table understanding-Saclay-H/F

CEA

Saclay

Presencial

EUR 8900 - 13.000

A tiempo parcial

14 días+

Consigue una respuesta de este empleador

Tu currículum y tu carta de presentación, adaptados exactamente a lo que buscan contratar.

Supera los filtros ATS

Descripción de la vacante

Le CEA, LIST/LASTI à Saclay, propose un stage en Reinforcement Learning appliqué aux VLMs pour la compréhension de tableaux. Vous rejoindrez une équipe de recherche axée NLP et vision par ordinateur, avec accès à un cluster GPU et à des ressources de pointe.

Le stage vise à explorer des signaux d’évaluation sans annotation humaine et à publier les résultats potentiellement. Le contexte comporte l’entraînement de VLMs et des expériences expérimentales, avec supervision par des chercheurs et

Formación

  • Étudiant en dernière année (M2 ou école d'ingénieurs).
  • Bon niveau d'anglais.
  • Connaissances en apprentissage profond et statistiques.
  • Expérience souhaitée avec HuggingFace Transformers.

Responsabilidades

  • Démarrer dans l’environnement d’entraînement et d’évaluation de l’équipe.
  • Analyser les méthodes d’évaluation automatiques.
  • Proposer et tester de nouvelles approches, évaluées par des expériences.
  • Améliorer les performances du pipeline et la robustesse des modèles.
  • Évaluer les modèles sur des benchmarks publics.
  • Documenter les résultats.
  • Travailler sur un pipeline existant avec accès à un cluster GPU (Python, PyTorch, HuggingFace, entraînement distribué, suivi des expériences).
  • Encadrement par des ingénieurs chercheurs, avec une réunion hebdomadaire.

Conocimientos

Python
PyTorch
Linux
Deep learning
Statistiques
Anglais

Educación

Master 2 ou école d'ingénieurs

Herramientas

HuggingFace Transformers
PyTorch
Linux

Descripción del empleo

Job title

Internship Reinforcement learning of VLMs for table understanding-Saclay-H/F

Contract

Internship

Subject

Based in Saclay (Essonne), the LIST is one of the two institutes of CEA Tech, the Technological Research Division of the CEA. Dedicated to intelligent digital systems, its mission is to carry out technological developments of excellence on behalf of industrial partners, in order to create value.
Within the LIST, the Laboratory of Textual and Visual Semantic Analysis (LASTI) conducts its research in the field of natural language processing and computer vision to extract, classify and generate information. The laboratory's research themes include learning with few data, trustworthiness and multimodality.

6 monhs

As an intern at the CEA, you will have the opportunity to work within a world-renowned research environment. Our teams consist of passionate and dedicated experts, providing an ideal setting for learning and collaboration. You will have access to state-of-the-art equipment and top-tier research resources to carry out your assignments. The work conducted may potentially lead to a scientific publication.

Context :

This internship focuses on training VLMs with reinforcement learning (GRPO), using reward signals that do not depend on human annotations: for example, checking the geometric consistencies or generating synthetic tables whose structure is known by construction. The central question is which annotation-free signals correlate well with the real quality of a prediction, and how to combine them to improve the model on real documents.

What do we expect from you?
  • Get started with the team's training and evaluation environment
  • Analyse automatic evaluation methods
  • Propose and implement new approaches, and validate them through experiments
  • Work on the performance of the pipeline and the robustness of the models
  • Evaluate the models on public benchmarks
  • Document the results
  • The intern works on an existing pipeline, with access to a GPU cluster (Python, PyTorch, HuggingFace ecosystem, distributed training, experiment tracking). Supervision research engineers of the team, with a weekly meeting.
Methods / Means
  • Students in their final year of studies (M2 or engineering school)
  • Python and PyTorch, Linux
  • Knowledge of deep learning and statistics
  • Good level of English
  • Appreciated: experience with HuggingFace Transformers, notions of reinforcement learning (GRPO, DPO, …)
Site

Saclay

Location

Saclay

Organisation

The French Alternative Energies and Atomic Energy Commission (CEA) is a key player in research, development and innovation in four main areas :

  • defence and security,
  • nuclear energy (fission and fusion),
  • technological research for industry,
  • fundamental research in the physical sciences and life sciences.

Drawing on its widely acknowledged expertise, and thanks to its 16000 technicians, engineers, researchers and staff, the CEA actively participates in collaborative projects with a large number of academic and industrial partners.

The CEA is established in ten centers spread throughout France

Reference

2026-42011

Consigue la evaluación confidencial y gratuita de tu currículum.

o arrastra y suelta tu archivo aquí

Similar jobs

Puestos de trabajo similares que vale la pena comparar

Intership Agentic Multimodal AI for Understanding and Reasoning over Complex Technical Documents-H/F
Intership Agentic Multimodal AI for Understanding and Reasoning over Complex Technical Documents-H/F

CEA • Saclay

Presencial
EUR 10.000 - 12.000
Internship in Artificial Intelligence-Saclay- H/F
Internship in Artificial Intelligence-Saclay- H/F

CEA • Saclay

Presencial
EUR 10.000 - 13.000
Python and PyTorch competency
Stage Human-Object Interaction benchmarking-Saclay-H/F
Stage Human-Object Interaction benchmarking-Saclay-H/F

CEA • Saclay

Presencial
EUR 12.000 - 16.000
Stage Steering LLM to inhibit biases- Saclay-H/F
Stage Steering LLM to inhibit biases- Saclay-H/F

Cea - Commissariat à l'Energie Atomique • Palaiseau

Presencial
EUR 11.000 - 15.000
Internship Generic correctness of iterative cirtical software architecture exploration-Saclay-H/F
Internship Generic correctness of iterative cirtical software architecture exploration-Saclay-H/F

CEA • Saclay

Presencial
EUR 14.000 - 17.000
Internship 3D Reconstruction in the Dark: Radiance Fields for Unlit Environments-Saclay-H/F
Internship 3D Reconstruction in the Dark: Radiance Fields for Unlit Environments-Saclay-H/F

CEA • Saclay

Presencial
EUR 10.000 - 17.000
Stage - Training networks to be compressible: low rank, quantization, and how they combine-Saclay-H/F
Stage - Training networks to be compressible: low rank, quantization, and how they combine-Saclay-H/F

Cea - Commissariat à l'Energie Atomique • Saclay

Presencial
EUR 12.000 - 17.000
Stage Case-Based Reasoning on Concepts for Sensible Explainable AI- Saclay-H/F
Stage Case-Based Reasoning on Concepts for Sensible Explainable AI- Saclay-H/F

CEA • Saclay

Presencial
EUR 12.000 - 17.000
State-of-the-art equipment
Publication potential
Internship in AI: Graph-based Artificial Neural Networks H/F - Grenoble
Internship in AI: Graph-based Artificial Neural Networks H/F - Grenoble

CEA • Grenoble

Presencial
EUR 11.000 - 13.000
Internship: RL for VLMs powering table understanding
Internship: RL for VLMs powering table understanding

CEA • Saclay

Presencial
EUR 8900 - 13.000