Publications by Karl Aberer | EPFL Graph Search

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

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

Karl Aberer, Thanh Trung Huynh

Federated learning (FL) has recently emerged as a compelling machine learning paradigm, prioritizing the protection of privacy for training data. The increasing demand to address issues such as "the right to be forgotten" and combat data poisoning attacks ...

Springer Nature2024

Message from the General Chairs and PC Chairs CIC 2024

Karl Aberer

Institute for Excellence and Ethics2024

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

Karl Aberer, Angeliki Romanou, Antoine Bosselut, Debjit Paul, Syrielle Mathilde Marie Montariol, Léo Jules Laugier

Understanding narratives requires reasoning about the cause-and-effect relationships between events mentioned in the text. While existing foundation models yield impressive results in many NLP tasks requiring reasoning, it is unclear whether they understan ...

Association for Computational Linguistics2023

Efficient Integration of Multi-Order Dynamics and Internal Dynamics in Stock Movement Prediction

Karl Aberer, Thanh Trung Huynh

Advances in deep neural network (DNN) architectures have enabled new prediction techniques for stock market data. Unlike other multivariate time-series data, stock markets show two unique characteristics: (i) multi-order dynamics, as stock prices are a.ect ...

Assoc Computing Machinery2023

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

Karl Aberer, Negar Foroutan Eghlidi, Mohammadreza Banaei, Antoine Bosselut

In this work, we study whether multilingual language models (MultiLMs) can transfer logical reasoning abilities to other languages when they are fine-tuned for reasoning in a different language. We evaluate the cross-lingual reasoning abilities of MultiLMs ...

Association for Computational Linguistics2023

Peer-Reviewing Processes and Incentives: Data Management Community Survey Results

Karl Aberer

Reviewing papers for conferences is an important and hard task that brings several challenges. The Data Management community has been increasingly struggling with high reviewer load, low-quality reviews and low reviewer engagement, unethical reviewing prac ...

Assoc Computing Machinery2023

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

Scalable maximal subgraph mining with backbone-preserving graph convolutions

Karl Aberer, Quoc Viet Hung Nguyen, Thành Tâm Nguyên, Thanh Trung Huynh

Maximal subgraph mining is increasingly important in various domains, including bioinformatics, genomics, and chemistry, as it helps identify common characteristics among a set of graphs and enables their classification into different categories. Existing ...

ELSEVIER SCIENCE INC2023

From scattered sources to comprehensive technology landscape : A recommendation-based retrieval approach

Karl Aberer, Chi Thang Duong

Mapping the technology landscape is crucial for market actors to take informed investment decisions. However, given the large amount of data on the Web and its subsequent information overload, manually retrieving information is a seemingly ineffective and ...

ELSEVIER2023

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

Preface-Special Issue on Misinformation on the Web

Karl Aberer, Quoc Viet Hung Nguyen

PERGAMON-ELSEVIER SCIENCE LTD2022

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, Negar Foroutan Eghlidi, Mohammadreza Banaei, Antoine Bosselut

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

Efficient and Effective Multi-Modal Queries Through Heterogeneous Network Embedding

Karl Aberer, Quoc Viet Hung Nguyen, Thành Tâm Nguyên, Chi Thang Duong, Thanh Trung Huynh

The heterogeneity of today's Web sources requires information retrieval (IR) systems to handle multi-modal queries. Such queries define a user's information needs by different data modalities, such as keywords, hashtags, user profiles, and other media. Rec ...

IEEE COMPUTER SOC2022

Nature vs. Nurture: Feature vs. Structure for Graph Neural Networks

Karl Aberer, Quoc Viet Hung Nguyen, Thành Tâm Nguyên, Chi Thang Duong

Graph neural networks take node features and graph structure as input to build representations for nodes and graphs. While there are a lot of focus on GNN models, understanding the impact of node features and graph structure to GNN performance has received ...

ELSEVIER2022

Scalable Robust Graph Embedding with Spark

Karl Aberer, Quoc Viet Hung Nguyen, Chi Thang Duong, Trung-Dung Hoang

Graph embedding aims at learning a vector-based representation of vertices that incorporates the structure of the graph. This representation then enables inference of graph properties. Existing graph embedding techniques, however, do not scale well to larg ...

ASSOC COMPUTING MACHINERY2021

Ephemeral Astroturfing Attacks: The Case of Fake Twitter Trends

Karl Aberer, Tugrulcan Elmas, Ahmed Furkan Ozkalay, Rebekah Jean Overdorf

We uncover a previously unknown, ongoing astroturfing attack on the popularity mechanisms of social media platforms: ephemeral astroturfing attacks. In this attack, a chosen keyword or topic is artificially promoted by coordinated and inauthentic activity ...

IEEE2021

Efficient Streaming Subgraph Isomorphism with Graph Neural Networks

Karl Aberer, Quoc Viet Hung Nguyen, Chi Thang Duong, Trung-Dung Hoang

Queries to detect isomorphic subgraphs are important in graph-based data management. While the problem of subgraph isomorphism search has received considerable attention for the static setting of a single query, or a batch thereof, existing approaches do n ...

ASSOC COMPUTING MACHINERY2021

SciClops: Detecting and Contextualizing Scientific Claims for Assisting Manual Fact-Checking

Karl Aberer, Panagiotis Smeros, Carlos Castillo

This paper describes SciClops, a method to help combat online scientific misinformation. Although automated fact-checking methods have gained significant attention recently, they require pre-existing ground-truth evidence, which, in the scientific context, ...

ACM2021