Publications

Ghazi Felhi, Joseph Le Roux, Djamé Seddah (2022). Exploiting Inductive Bias in Transformers for Unsupervised Disentanglement of Syntax and Semantics with VAEs. NAACL.

PDF Code Slides Video

Ghazi Felhi, Joseph Le Roux, Djamé Seddah (2021). Towards Unsupervised Content Disentanglement in Sentence Representations via Syntactic Roles. CtrlGen@NeurIPS21.

PDF Code Poster

Ghazi Felhi, Joseph Le Roux, Djamé Seddah (2021). Challenging the Semi-Supervised VAE Framework for Text Classification. Insights@EMNLP21.

PDF Code Poster Video DOI

Ghazi Felhi, Joseph Le Roux, Djamé Seddah (2020). Disentangling semantics in language through VAEs and a certain architectural choice. Arxiv.

Ghazi Felhi, Joseph Le Roux, Djamé Seddah (2020). Controlling the Interaction Between Generation and Inference in Semi-Supervised Variational Autoencoders Using Importance Weighting. Arxiv.

Davide Buscaldi, Ghazi Felhi, Dhaou Ghoul, Joseph LeRoux, Gaël Lejeune, Xudong Zhang (2020). Calcul de similarité entre phrases : quelles mesures et quels descripteurs ?. JEP/TALN/RECITAL.

Léo Hemamou, Ghazi Felhi, Jean-Claude Martin, Chloé Clavel (2019). Slices of Attention in Asynchronous Video Job Interviews. ACII.

Loïc Coquelin, Nicolas Fischer, Nicolas Feltin, Laurent Devoille, Ghazi Felhi (2019). Towards the use of deep generative models for the characterization in size of aggregated TiO2 nanoparticles measured by Scanning Electron Microscopy (SEM). Materials Research Express.

Léo Hemamou, Ghazi Felhi, Vincent Vandenbussche, Jean-Claude Martin, Chloé Clavel (2019). HireNet: A Hierarchical Attention Model for the Automatic Analysis of Asynchronous Video Job Interviews. AAAI.