Search Results for author: Michael Granitzer

Found 51 papers, 24 papers with code

GRhOOT: Ontology of Rhetorical Figures in German

1 code implementation LREC 2022 Ramona Kühn, Jelena Mitrović, Michael Granitzer

GRhOOT, the German RhetOrical OnTology, is a domain ontology of 110 rhetorical figures in the German language.

Argument Mining Machine Translation +1

Bridging the Semantic Gap in Virtual Machine Introspection and Forensic Memory Analysis

no code implementations7 Mar 2025 Christofer Fellicious, Hans P. Reiser, Michael Granitzer

We also test our method on complete physical memory dumps to showcase the effectiveness of the engineered features.

Feature Engineering

WebFAQ: A Multilingual Collection of Natural Q&A Datasets for Dense Retrieval

no code implementations28 Feb 2025 Michael Dinzinger, Laura Caspari, Kanishka Ghosh Dastidar, Jelena Mitrović, Michael Granitzer

These datasets are carefully curated through refined filtering and near-duplicate detection, yielding high-quality resources for training and evaluating multilingual dense retrieval models.

Dataset Generation Open-Domain Question Answering +2

Malware Detection based on API calls

1 code implementation18 Feb 2025 Christofer Fellicious, Manuel Bischof, Kevin Mayer, Dorian Eikenberg, Stefan Hausotte, Hans P. Reiser, Michael Granitzer

We also empirically show that we only need a subset of the function call sequence, specifically calls to the ntdll. dll library, to identify malware.

Malware Detection

On the Suitability of pre-trained foundational LLMs for Analysis in German Legal Education

no code implementations20 Dec 2024 Lorenz Wendlinger, Christian Braun, Abdullah Al Zubaer, Simon Alexander Nonn, Sarah Großkopf, Christofer Fellicious, Michael Granitzer

We show that current open-source foundational LLMs possess instruction capability and German legal background knowledge that is sufficient for some legal analysis in an educational context.

Argument Mining Automated Essay Scoring +1

Enhancing Rhetorical Figure Annotation: An Ontology-Based Web Application with RAG Integration

1 code implementation18 Dec 2024 Ramona Kühn, Jelena Mitrović, Michael Granitzer

To overcome this issue, we develop a web application called "Find your Figure" that facilitates the identification and annotation of German rhetorical figures.

Argument Mining Fake News Detection +4

Krony-PT: GPT2 compressed with Kronecker Products

no code implementations16 Dec 2024 M. Ayoub Ben Ayad, Jelena Mitrovic, Michael Granitzer

We introduce Krony-PT, a compression technique of GPT2 \citep{radford2019language} based on Kronecker Products.

Language Modeling Language Modelling

SUDS: A Strategy for Unsupervised Drift Sampling

1 code implementation5 Nov 2024 Christofer Fellicious, Lorenz Wendlinger, Mario Gancarski, Jelena Mitrovic, Michael Granitzer

We also introduce the Harmonized Annotated Data Accuracy Metric (HADAM), a metric that evaluates classifier performance in relation to the quantity of annotated data required to achieve the stated performance, thereby taking into account the difficulty of acquiring labeled data.

Drift Detection

Towards an Improved Metric for Evaluating Disentangled Representations

1 code implementation4 Oct 2024 Sahib Julka, Yashu Wang, Michael Granitzer

Disentangled representation learning plays a pivotal role in making representations controllable, interpretable and transferable.

Disentanglement

Mixture of Modular Experts: Distilling Knowledge from a Multilingual Teacher into Specialized Modular Language Models

1 code implementation28 Jul 2024 Mohammed Al-Maamari, Mehdi Ben Amor, Michael Granitzer

Evaluations of modular MoE architectures revealed that Pre-trained Language Experts (PLE) and Joint Expert Embedding Training (JEET) performed similarly, while the MoE with Common Expert (MoE-CE) setup showed slightly lower performance.

Knowledge Distillation Mixture-of-Experts

PersonaRAG: Enhancing Retrieval-Augmented Generation Systems with User-Centric Agents

1 code implementation12 Jul 2024 Saber Zerhoudi, Michael Granitzer

Large Language Models (LLMs) struggle with generating reliable outputs due to outdated knowledge and hallucinations.

Information Retrieval Question Answering +3

DriftGAN: Using historical data for Unsupervised Recurring Drift Detection

1 code implementation9 Jul 2024 Christofer Fellicious, Sahib Julka, Lorenz Wendlinger, Michael Granitzer

In real-world applications, input data distributions are rarely static over a period of time, a phenomenon known as concept drift.

Drift Detection

Challenges and Considerations in Annotating Legal Data: A Comprehensive Overview

no code implementations5 Jul 2024 Harshil Darji, Jelena Mitrović, Michael Granitzer

This paper provides an expanded view of these challenges and aims to offer a foundational understanding and guidance for researchers and professionals engaged in legal data annotation projects.

Computational Approaches to the Detection of Lesser-Known Rhetorical Figures: A Systematic Survey and Research Challenges

no code implementations24 Jun 2024 Ramona Kühn, Jelena Mitrović, Michael Granitzer

Rhetorical figures play a major role in our everyday communication as they make text more interesting, more memorable, or more persuasive.

Efficient NAS with FaDE on Hierarchical Spaces

no code implementations24 Apr 2024 Simon Neumeyer, Julian Stier, Michael Granitzer

Our experiments show that firstly, FaDE-ranks on finite regions of the search space correlate with corresponding architecture performances and secondly, the ranks can empower a pseudo-gradient evolutionary search on the complete neural architecture search space.

Neural Architecture Search

A Survey of Web Content Control for Generative AI

no code implementations2 Apr 2024 Michael Dinzinger, Florian Heß, Michael Granitzer

The groundbreaking advancements around generative AI have recently caused a wave of concern culminating in a row of lawsuits, including high-profile actions against Stability AI and OpenAI.

Survey

Towards Measuring Representational Similarity of Large Language Models

1 code implementation5 Dec 2023 Max Klabunde, Mehdi Ben Amor, Michael Granitzer, Florian Lemmerich

Understanding the similarity of the numerous released large language models (LLMs) has many uses, e. g., simplifying model selection, detecting illegal model reuse, and advancing our understanding of what makes LLMs perform well.

Model Selection

Is GPT-4 a reliable rater? Evaluating Consistency in GPT-4 Text Ratings

no code implementations3 Aug 2023 Veronika Hackl, Alexandra Elena Müller, Michael Granitzer, Maximilian Sailer

Statistical analysis was conducted in order to learn more about the interrater reliability, consistency of the ratings across iterations and the correlation between ratings in terms of content and style.

Language Modeling Language Modelling +1

GRAN is superior to GraphRNN: node orderings, kernel- and graph embeddings-based metrics for graph generators

1 code implementation13 Jul 2023 Ousmane Touat, Julian Stier, Pierre-Edouard Portier, Michael Granitzer

We use these metrics to compare GraphRNN and GRAN, two well-known generative models for graphs, and unveil the influence of node orderings.

Drug Discovery Graph Embedding +2

Knowledge distillation with Segment Anything (SAM) model for Planetary Geological Mapping

no code implementations12 May 2023 Sahib Julka, Michael Granitzer

Planetary science research involves analysing vast amounts of remote sensing data, which are often costly and time-consuming to annotate and process.

Decoder Image Segmentation +2

Technical Report: Impact of Position Bias on Language Models in Token Classification

1 code implementation26 Apr 2023 Mehdi Ben Amor, Michael Granitzer, Jelena Mitrović

Therefore, we conduct an in-depth evaluation of the impact of position bias on the performance of LMs when fine-tuned on token classification benchmarks.

named-entity-recognition Named Entity Recognition +7

German BERT Model for Legal Named Entity Recognition

no code implementations7 Mar 2023 Harshil Darji, Jelena Mitrović, Michael Granitzer

Even though there is much research done on NER using BERT and other popular language models, the same is not explored in detail when it comes to Legal NLP or Legal Tech.

Language Modelling model +5

SmartKex: Machine Learning Assisted SSH Keys Extraction From The Heap Dump

1 code implementation12 Sep 2022 Christofer Fellicious, Stewart Sentanoe, Michael Granitzer, Hans P. Reiser

A commonly used method in digital forensics is to extract data from the main memory of a digital device.

The Importance of Future Information in Credit Card Fraud Detection

no code implementations11 Apr 2022 Van Bach Nguyen, Kanishka Ghosh Dastidar, Michael Granitzer, Wissam Siblini

We believe that future works on this new paradigm will have a significant impact on the detection of compromised cards.

Fraud Detection

deepstruct -- linking deep learning and graph theory

1 code implementation12 Nov 2021 Julian Stier, Michael Granitzer

deepstruct connects deep learning models and graph theory such that different graph structures can be imposed on neural networks or graph structures can be extracted from trained neural network models.

Deep Learning Neural Architecture Search

Experiments on Properties of Hidden Structures of Sparse Neural Networks

1 code implementation27 Jul 2021 Julian Stier, Harshil Darji, Michael Granitzer

Sparsity in the structure of Neural Networks can lead to less energy consumption, less memory usage, faster computation times on convenient hardware, and automated machine learning.

Neural Architecture Search Prediction

nlpUP at SemEval-2020 Task 12 : A Blazing Fast System for Offensive Language Detection

no code implementations SEMEVAL 2020 Ehab Hamdy, Jelena Mitrovi{\'c}, Michael Granitzer

In this paper, we introduce our submission for the SemEval Task 12, sub-tasks A and B for offensive language identification and categorization in English tweets.

Language Identification

ADSAGE: Anomaly Detection in Sequences of Attributed Graph Edges applied to insider threat detection at fine-grained level

no code implementations14 Jul 2020 Mathieu Garchery, Michael Granitzer

We evaluate ADSAGE on authentication, email traffic and web browsing logs from the CERT insider threat datasets, as well as on real-world authentication events.

Anomaly Detection Feature Engineering

DeepGG: a Deep Graph Generator

1 code implementation7 Jun 2020 Julian Stier, Michael Granitzer

Learning distributions of graphs can be used for automatic drug discovery, molecular design, complex network analysis, and much more.

Drug Discovery Graph Embedding

Language Proficiency Scoring

no code implementations LREC 2020 Cristina Arhiliuc, Jelena Mitrovi{\'c}, Michael Granitzer

The Common European Framework of Reference (CEFR) provides generic guidelines for the evaluation of language proficiency.

Investigating Extensions to Random Walk Based Graph Embedding

no code implementations17 Feb 2020 Joerg Schloetterer, Martin Wehking, Fatemeh Salehi Rizi, Michael Granitzer

Graph embedding has recently gained momentum in the research community, in particular after the introduction of random walk and neural network based approaches.

Graph Embedding Link Prediction +1

Predicting event attendance exploring social influence

no code implementations16 Feb 2020 Fatemeh Salehi Rizi, Michael Granitzer

In this paper, we propose to model the social influence of friends on event attendance.

Graph Embedding

Global and Local Feature Learning for Ego-Network Analysis

no code implementations16 Feb 2020 Fatemeh Salehi Rizi, Michael Granitzer, Konstantin Ziegler

This social network can be efficiently analyzed after learning representations of the ego and its alters in a low-dimensional, real vector space.

Language Modeling Language Modelling +1

Shortest path distance approximation using deep learning techniques

1 code implementation12 Feb 2020 Fatemeh Salehi Rizi, Joerg Schloetterer, Michael Granitzer

Computing shortest path distances between nodes lies at the heart of many graph algorithms and applications.

Deep Learning

Competitive Influence Maximization: Integrating Budget Allocation and Seed Selection

1 code implementation27 Dec 2019 Amirhossein Ansari, Masoud Dadgar, Ali Hamzeh, Jörg Schlötterer, Michael Granitzer

In this work, we integrate these two lines of research and propose a new scenario where competition happens in two phases.

Social and Information Networks Computer Science and Game Theory

Parallel Total Variation Distance Estimation with Neural Networks for Merging Over-Clusterings

no code implementations9 Dec 2019 Christian Reiser, Jörg Schlötterer, Michael Granitzer

We consider the initial situation where a dataset has been over-partitioned into $k$ clusters and seek a domain independent way to merge those initial clusters.

Policy Learning for Malaria Control

2 code implementations20 Oct 2019 Van Bach Nguyen, Belaid Mohamed Karim, Bao Long Vu, Jörg Schlötterer, Michael Granitzer

In this report, we introduce the progress to learn the policy for Malaria Control as a Reinforcement Learning problem in the KDD Cup Challenge 2019 and propose diverse solutions to deal with the limited observations problem.

Bayesian Optimization Decision Making +5

Structural Analysis of Sparse Neural Networks

no code implementations16 Oct 2019 Julian Stier, Michael Granitzer

Sparse Neural Networks regained attention due to their potential for mathematical and computational advantages.

Neural Architecture Search

Towards automated feature engineering for credit card fraud detection using multi-perspective HMMs

1 code implementation3 Sep 2019 Yvan Lucas, Pierre-Edouard Portier, Léa Laporte, Liyun He-Guelton, Olivier Caelen, Michael Granitzer, Sylvie Calabretto

Our multiple perspectives HMM-based approach offers automated feature engineering to model temporal correlations so as to improve the effectiveness of the classification task and allows for an increase in the detection of fraudulent transactions when combined with the state of the art expert based feature engineering strategy for credit card fraud detection.

Automated Feature Engineering Feature Engineering +2

nlpUP at SemEval-2019 Task 6: A Deep Neural Language Model for Offensive Language Detection

no code implementations SEMEVAL 2019 Jelena Mitrovi{\'c}, Bastian Birkeneder, Michael Granitzer

In addition, we evaluate our approach on a different dataset and show that our model is capable of detecting online aggressiveness in both English and German tweets.

Language Modeling Language Modelling

Multiple perspectives HMM-based feature engineering for credit card fraud detection

1 code implementation15 May 2019 Yvan Lucas, Pierre-Edouard Portier, Léa Laporte, Olivier Caelen, Liyun He-Guelton, Sylvie Calabretto, Michael Granitzer

In this article, we model a sequence of credit card transactions from three different perspectives, namely (i) does the sequence contain a Fraud?

Feature Engineering Fraud Detection

Analysing Neural Network Topologies: a Game Theoretic Approach

no code implementations17 Apr 2019 Julian Stier, Gabriele Gianini, Michael Granitzer, Konstantin Ziegler

In previous work, heuristics based on using the weight distribution of a neuron as contribution measure have shown some success, but do not provide a proper theoretical understanding.

Network Pruning

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