Publications by Rémi Philippe Lebret

SAGESSE: A System for Argument Generation, Extraction and Structuring of Social Exchanges

Karl Aberer, Rémi Philippe Lebret, Léo Jules Laugier

Online debates provide critical insights into public opinion and societal trends, yet the unstructured nature of these discussions presents significant challenges for analysis. In this paper, we present SAGESSE, a novel argumentation parsing pipeline tailo ...

ACM2025

Stop Pre-Training: Adapt Visual-Language Models to Unseen Languages

Karl Aberer, Rémi Philippe Lebret, Negar Foroutan Eghlidi

Vision-Language Pre-training (VLP) has advanced the performance of many visionlanguage tasks, such as image-text retrieval, visual entailment, and visual reasoning. The pre-training mostly utilizes lexical databases and image queries in English. Previous w ...

Assoc Computational Linguistics-Acl2023

Revisiting Offline Compression: Going Beyond Factorization-based Methods for Transformer Language Models

Karl Aberer, Rémi Philippe Lebret, Mohammadreza Banaei

Recent transformer language models achieve outstanding results in many natural language processing (NLP) tasks. However, their enormous size often makes them impractical on memory-constrained devices, requiring practitioners to compress them to smaller net ...

Assoc Computational Linguistics-Acl2023

MoZuMa: A Model Zoo for Multimedia Applications

Karl Aberer, Rémi Philippe Lebret

Lots of machine learning models with applications in Multimedia Search are released as Open Source Software. However, integrating these models into an application is not always an easy task due to the lack of a consistent interface to run, train or distrib ...

Assoc Computing Machinery2022

Discovering Language-neutral Sub-networks in Multilingual Language Models

Karl Aberer, Rémi Philippe Lebret, Antoine Bosselut, Negar Foroutan Eghlidi, Mohammadreza Banaei

Multilingual pre-trained language models transfer remarkably well on cross-lingual downstream tasks. However, the extent to which they learn language-neutral representations (i.e., shared representations that encode similar phenomena across languages), and ...

Association for Computational Linguistics2022

Direction is what you need: Improving Word Embedding Compression in Large Language Models

Karl Aberer, Rémi Philippe Lebret, Mohammadreza Banaei

The adoption of Transformer-based models in natural language processing (NLP) has led to great success using a massive number of parameters. However, due to deployment constraints in edge devices, there has been a rising interest in the compression of thes ...

ASSOC COMPUTATIONAL LINGUISTICS-ACL2021

Upgrading the Newsroom: An Automated Image Selection System for News Articles

Karl Aberer, Rémi Philippe Lebret, Fangyu Liu

We propose an automated image selection system to assist photo editors in selecting suitable images for news articles. The system fuses multiple textual sources extracted from news articles and accepts multilingual inputs. It is equipped with char-level wo ...

ASSOC COMPUTING MACHINERY2020

Aligning Multilingual Word Embeddings for Cross-Modal Retrieval Task

Karl Aberer, Rémi Philippe Lebret, Alireza Mohammadshahi

In this paper, we propose a new approach to learn multimodal multilingual embeddings for matching images and their relevant captions in two languages. We combine two existing objective functions to make images and captions close in a joint embedding space ...

Association for Computational Linguistics2019

Polisis: Automated Analysis and Presentation of Privacy Policies Using Deep Learning

Karl Aberer, Rémi Philippe Lebret, Hamza Harkous, Kassem Fawaz

Privacy policies are the primary channel through which companies inform users about their data collection and sharing practices. These policies are often long and difficult to comprehend. Short notices based on information extracted from privacy policies h ...

USENIX ASSOC2018

280 Birds with One Stone: Inducing Multilingual Taxonomies from Wikipedia Using Character-Level Classification

Karl Aberer, Rémi Philippe Lebret, Hamza Harkous, Amit Gupta

We propose a novel fully-automated approach towards inducing multilingual taxonomies fromWikipedia. Given an English taxonomy, our approach first leverages the interlanguage links of Wikipedia to automatically construct training datasets for the is-a relat ...

ASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE2018