My projects

• Science communication
• Research

—

Science communication

Puppets of a Digital Brain – a (not so) short film about the technology behind generative AI, and its good and bad sides.

As member of SciFilmIt, together with the Associatia Persona from Cluj-Napoca (Romania), we organized two events for making short films in 2024:

• In the workshop in April a group of pupils from the National Coolege George Cosbuc in Cluj-Napoca made very fun and informative short films on 5 topics, as an active learning activity.
• In the workshop in December, we had participants form three universities in Cluj-Napoca -- the University of Agricultural Sciences and Veterinaryr Medicine (USAMV), University Babes-Bolyai (UBB), The National Music Academy "Gheorghe Dima" (ANMGD) -- who combined their expertise to created short films about the research at USAMV

Research

NCCR Evolving language 2021 –

• Disentangling text representations
• Animal call segmentation -- a variation on voxseg (with more parameters)
• Word change tracking over time -- an updated version of histwords
• Analysis of case syncretism patterns using UniMorph

University of Heidelberg/HITS/FBK/University of Stuttgart

• Semantic relations
  • Semantic Relations between Nominals
  • Semantic Relations and Deep Learning
  • Semantic Relations Across Syntactic Levels
  • The same semantic relations link structurally different realizations of concepts
  • Classification of semantic relations between nominals
  • Semeval-2007 task 04: Classification of semantic relations between nominals
  • Learning noun-modifier semantic relations with corpus-based and WordNet-based features
  • Exploring noun-modifier semantic relations
  • Unifying semantic relations across syntactic levels
• Growing and abstracting knowledge graphs
  • Abstract graphs and abstract paths for knowledge base completion
  • Using patterns in knowledge graphs for targeted information extraction
  • Analysis of the Impact of Negative Sampling on Link Prediction in Knowledge Graphs
  • Learning Knowledge Graph Embeddings with Type Regularizer
  • Automatic Reconstruction of Emperor Itineraries from the Regesta Imperii
  • A Knowledge Graph from the Regesta Imperii: Construction, Visualization and Macro-level Analyses
  • Induction of a Large-Scale Knowledge Graph from the Regesta Imperii
  </ul>
• Wikipedia as a knowledge source
  • Transforming Wikipedia into a large scale multilingual concept network
  • Decoding Wikipedia Categories for Knowledge Acquisition
  • WikiNet: A Very Large Scale Multi-Lingual Concept Network
  • Concept-based selectional preferences and distributional representations from wikipedia articles
  • knoWitiary: A Machine Readable Incarnation of Wiktionary
  • Coarse-grained Cross-lingual Alignment of Comparable Texts with Topic Models and Encyclopedic Knowledge
  • Mapping WordNet Domains, WordNet Topics and Wikipedia Categories to Generate Multilingual Domain Specific Resources
  • Topic-driven multi-document summarization with encyclopedic knowledge and spreading activation
• Figurative language, word senses and language change
  • Word epoch disambiguation: Finding how words change over time
  • Metaphors in Text Simplification: To change or not to change, that is the question
  • Local and global context for supervised and unsupervised metonymy resolution
  • Combining collocations, lexical and encyclopedic knowledge for metonymy resolution
  • Bridging languages through etymology: The case of cross language text categorization
  • Unsupervised All-words Word Sense Disambiguation with Grammatical Dependencies
  • Word sense disambiguation in Roget's thesaurus using WordNet
  • What's in a name? In some languages, grammatical gender
  • Do happy words sound happy? a study of the relation between form and meaning for english words expressing emotions
• Argumentation mining
  • Towards Explaining Natural Language Arguments with Background Knowledge
  • Enriching Argumentative Texts with Implicit Knowledge
  • Classifying Semantic Clause Types: Modeling Context and Genre Characteristics with Recurrent Neural Networks and Attention
• Native language interference
  • Exploiting native language interference for native language identification
  • Anglicized Words and Misspelled Cognates in Native Language Identification
  • The Role of Emotions in Native Language Identification
  • Punctuation as Native Language Interference
  • Word Etymology as Native Language Interference
  • Improving native language identification by using spelling errors
• Correcting OCR errors using MT methods: Correction of OCR Word Segmentation Errors in Articles from the ACL Collection through Neural Machine Translation Methods