Search Results for author: Karl Aberer

Found 32 papers, 11 papers with code

Multilingual Text Summarization on Financial Documents

no code implementations FNP (LREC) 2022 Negar Foroutan, Angelika Romanou, Stéphane Massonnet, Rémi Lebret, Karl Aberer

The language models were fine-tuned on a financial document collection of three languages (English, Spanish, and Greek) and aim to identify the beginning of the summary narrative part of the document.

Abstractive Text Summarization

Fast-FedUL: A Training-Free Federated Unlearning with Provable Skew Resilience

1 code implementation28 May 2024 Thanh Trung Huynh, Trong Bang Nguyen, Phi Le Nguyen, Thanh Tam Nguyen, Matthias Weidlich, Quoc Viet Hung Nguyen, Karl Aberer

Federated learning (FL) has recently emerged as a compelling machine learning paradigm, prioritizing the protection of privacy for training data.

Backdoor Attack Data Poisoning +1

LoRA-XS: Low-Rank Adaptation with Extremely Small Number of Parameters

1 code implementation27 May 2024 Klaudia Bałazy, Mohammadreza Banaei, Karl Aberer, Jacek Tabor

The recent trend in scaling language models has led to a growing demand for parameter-efficient tuning (PEFT) methods such as LoRA (Low-Rank Adaptation).

Benchmarking GSM8K +1

Stance Detection on Social Media with Fine-Tuned Large Language Models

no code implementations18 Apr 2024 İlker Gül, Rémi Lebret, Karl Aberer

This study emphasizes the potential of LLMs in stance detection and calls for more extensive research in this field.

Few-Shot Learning Stance Detection

CRAB: Assessing the Strength of Causal Relationships Between Real-world Events

1 code implementation7 Nov 2023 Angelika Romanou, Syrielle Montariol, Debjit Paul, Leo Laugier, Karl Aberer, Antoine Bosselut

In this work, we present CRAB, a new Causal Reasoning Assessment Benchmark designed to evaluate causal understanding of events in real-world narratives.

Breaking the Language Barrier: Improving Cross-Lingual Reasoning with Structured Self-Attention

1 code implementation23 Oct 2023 Negar Foroutan, Mohammadreza Banaei, Karl Aberer, Antoine Bosselut

We evaluate the cross-lingual reasoning abilities of MultiLMs in two schemes: (1) where the language of the context and the question remain the same in the new languages that are tested (i. e., the reasoning is still monolingual, but the model must transfer the learned reasoning ability across languages), and (2) where the language of the context and the question is different (which we term code-switched reasoning).

Logical Reasoning

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

1 code implementation29 Jun 2023 Yasmine Karoui, Rémi Lebret, Negar Foroutan, Karl Aberer

Our evaluation across three distinct tasks (image-text retrieval, visual entailment, and natural language visual reasoning) demonstrates that this approach outperforms the state-of-the-art multilingual vision-language models without requiring large parallel corpora.

Machine Translation Retrieval +3

From Scattered Sources to Comprehensive Technology Landscape: A Recommendation-based Retrieval Approach

no code implementations9 Dec 2021 Chi Thang Duong, Dimitri Percia David, Ljiljana Dolamic, Alain Mermoud, Vincent Lenders, Karl Aberer

This is a two-task setup involving (i) technology classification of entities extracted from company corpus, and (ii) technology and company retrieval based on classified technologies.

Language Modelling Retrieval

Legal Transformer Models May Not Always Help

no code implementations14 Sep 2021 Saibo Geng, Rémi Lebret, Karl Aberer

This work investigates the value of domain adaptive pre-training and language adapters in legal NLP tasks.

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

1 code implementation ACL (RepL4NLP) 2021 Klaudia Bałazy, Mohammadreza Banaei, Rémi Lebret, Jacek Tabor, Karl Aberer

The adoption of Transformer-based models in natural language processing (NLP) has led to great success using a massive number of parameters.

Language Modelling

On Representation Learning for Scientific News Articles Using Heterogeneous Knowledge Graphs

no code implementations12 Apr 2021 Angelika Romanou, Panayiotis Smeros, Karl Aberer

In this work, we present a methodology for creating scientific news article representations by modeling the directed graph between the scientific news articles and the cited scientific publications.

Fact Checking Graph Neural Network +5

SciLens News Platform: A System for Real-Time Evaluation of News Articles

no code implementations27 Aug 2020 Angelika Romanou, Panayiotis Smeros, Carlos Castillo, Karl Aberer

We demonstrate the SciLens News Platform, a novel system for evaluating the quality of news articles.

Spoken dialect identification in Twitter using a multi-filter architecture

no code implementations5 Jun 2020 Mohammadreza Banaei, Rémi Lebret, Karl Aberer

This paper presents our approach for SwissText & KONVENS 2020 shared task 2, which is a multi-stage neural model for Swiss German (GSW) identification on Twitter.

Dialect Identification Task 2

Aligning Multilingual Word Embeddings for Cross-Modal Retrieval Task

1 code implementation EMNLP (WS) 2019 Alireza Mohammadshahi, Remi Lebret, Karl Aberer

In this paper, we propose a new approach to learn multimodal multilingual embeddings for matching images and their relevant captions in two languages.

Cross-Modal Retrieval Image-to-Text Retrieval +3

Unsupervised Learning of Node Embeddings by Detecting Communities

no code implementations25 Sep 2019 Chi Thang Duong, Dung Hoang, Truong Giang Le Ba, Thanh Le Cong, Hongzhi Yin, Matthias Weidlich, Quoc Viet Hung Nguyen, Karl Aberer

We provide empirical evidence that the communities learned by DMC are meaningful and that the node embeddings are competitive in different node classification benchmarks.

Clustering Node Classification +1

Parallel Computation of Graph Embeddings

no code implementations6 Sep 2019 Chi Thang Duong, Hongzhi Yin, Thanh Dat Hoang, Truong Giang Le Ba, Matthias Weidlich, Quoc Viet Hung Nguyen, Karl Aberer

We therefore propose a framework for parallel computation of a graph embedding using a cluster of compute nodes with resource constraints.

Graph Embedding

Weakly Supervised Active Learning with Cluster Annotation

no code implementations31 Dec 2018 Fábio Perez, Rémi Lebret, Karl Aberer

In this work, we introduce a novel framework that employs cluster annotation to boost active learning by reducing the number of human interactions required to train deep neural networks.

Active Learning

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

2 code implementations7 Feb 2018 Hamza Harkous, Kassem Fawaz, Rémi Lebret, Florian Schaub, Kang G. Shin, Karl Aberer

Companies, users, researchers, and regulators still lack usable and scalable tools to cope with the breadth and depth of privacy policies.

Language Modelling Question Answering

Multimodal Classification for Analysing Social Media

no code implementations7 Aug 2017 Chi Thang Duong, Remi Lebret, Karl Aberer

Although information on social media can be of different modalities such as texts, images, audio or videos, traditional approaches in classification usually leverage only one prominent modality.

Auxiliary Learning Classification +2

Taxonomy Induction using Hypernym Subsequences

no code implementations25 Apr 2017 Amit Gupta, Rémi Lebret, Hamza Harkous, Karl Aberer

We propose a novel, semi-supervised approach towards domain taxonomy induction from an input vocabulary of seed terms.

Robust Online Time Series Prediction with Recurrent Neural Networks

no code implementations26 Dec 2016 Tian Guo, Zhao Xu, Xin Yao, Haifeng Chen, Karl Aberer, Koichi Funaya

Time series forecasting for streaming data plays an important role in many real applications, ranging from IoT systems, cyber-networks, to industrial systems and healthcare.

Time Series Time Series Forecasting +1

Matching Demand with Supply in the Smart Grid using Agent-Based Multiunit Auction

no code implementations22 Aug 2013 Tri Kurniawan Wijaya, Kate Larson, Karl Aberer

Recent work has suggested reducing electricity generation cost by cutting the peak to average ratio (PAR) without reducing the total amount of the loads.

Cannot find the paper you are looking for? You can Submit a new open access paper.