GRISEZ, Mme Laure (2025) NLP for Structured data and inversion of physics oriented problems through sampling PFE - Project Graduation, ENSTA.
Full text not available from this repository.
Abstract
This report presents the work performed during my end-of-studies internship at Headmind Partners, it is divided into two parts. The first part deals with the subject of natural language processing (NLP) applied to the analysis and synthesis of complex tabular data. In this context, two models based on LLMs were designed and the different architectures compared. The second part focuses on my work carried out during my mission at TotalEnergies. This work concentrates on the application of Bayesian sampling methods (MCMC and Hamiltonian Monte Carlo) to geological model inversion
| Item Type: | Thesis (PFE - Project Graduation) |
|---|---|
| Uncontrolled Keywords: | LLMs, Agents, NLP, Monte Carlo Markov Chain, sampling, inversion problem |
| Subjects: | Information and Communication Sciences and Technologies Mathematics and Applications |
| ID Code: | 10837 |
| Deposited By: | Laure GRISEZ |
| Deposited On: | 08 oct. 2025 10:39 |
| Dernière modification: | 08 oct. 2025 10:39 |
Repository Staff Only: item control page
