Liviu P. Dinu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Liviu P. Dinu has a total of 43 co-authors. Here are the top co-authors by number of shared publications:
- Ana-Maria Bucur
Interdisciplinary School of Doctoral Studies, University of Bucharest, Bucharest, Romania
8 shared publications - Ana Sabina Uban
Faculty of Mathematics and Computer Science University of Bucharest
6 shared publications - Claudiu Creanga
Interdisciplinary School of Doctoral Studies
6 shared publications - Adrian Cosma
Faculty of Automatic Control and Computer Science University Politehnica of Bucharest
4 shared publications Alina Maria Ciobanu
University of Bucharest, Department of Computer Science, HLT Research Center
4 shared publications- Andrea Sgarro
University of Trieste
3 shared publications
Latest Publications
- Open Access
Measuring cross-language intelligibility between Romance languages with computational tools
We present an analysis of mutual intelligibility in related languages applied for languages in the Romance family. We introduce a novel computational metric for estimating intelligibility based on lexical similarity using surface and semantic similarity of related words, and use it to measure mutual intelligibility for the five main Romance languages (French, Italian, Portuguese, Spanish, and Romanian), and compare results using both the orthographic and phonetic forms of words as well as different parallel corpora and vectorial models of word meaning representation. The obtained intelligibility scores confirm intuitions related to intelligibility asymmetry across languages and significantly correlate with results of cloze tests in human experiments.
Open AccessMeasuring cross-language intelligibility between Romance languages with computational tools
We present an analysis of mutual intelligibility in related languages applied for languages in the Romance family. We introduce a novel computational metric for estimating intelligibility based on lexical similarity using surface and semantic similarity of related words, and use it to measure mutual intelligibility for the five main Romance languages (French, Italian, Portuguese, Spanish, and Romanian), and compare results using both the orthographic and phonetic forms of words as well as different parallel corpora and vectorial models of word meaning representation. The obtained intelligibility scores confirm intuitions related to intelligibility asymmetry across languages and significantly correlate with results of cloze tests in human experiments.
- Open Access
Few-shot text-based emotion detection
This paper describes the approach of the Unibuc - NLP team in tackling the SemEval 2025 Workshop, Task 11: Bridging the Gap in Text-Based Emotion Detection. We mainly focused on experiments using large language models (Gemini, Qwen, DeepSeek) with either few-shot prompting or fine-tuning. With our final system, for the multi-label emotion detection track (track A), we got an F1-macro of $0.7546$ (26/96 teams) for the English subset, $0.1727$ (35/36 teams) for the Portuguese (Mozambican) subset and $0.325$ (\textbf{1}/31 teams) for the Emakhuwa subset.Open AccessFew-shot text-based emotion detection
This paper describes the approach of the Unibuc - NLP team in tackling the SemEval 2025 Workshop, Task 11: Bridging the Gap in Text-Based Emotion Detection. We mainly focused on experiments using large language models (Gemini, Qwen, DeepSeek) with either few-shot prompting or fine-tuning. With our final system, for the multi-label emotion detection track (track A), we got an F1-macro of $0.7546$ (26/96 teams) for the English subset, $0.1727$ (35/36 teams) for the Portuguese (Mozambican) subset and $0.325$ (\textbf{1}/31 teams) for the Emakhuwa subset. - Open Access
- 25 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 450-457
Evaluating Large Language Models’ ability to pastiche literary texts
- 25 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 450-457
Open AccessEvaluating Large Language Models’ ability to pastiche literary texts
- Open Access
- 25 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 486-491
Unmasking Propaganda: A Comparative Analysis of Masked and Causal Language Models
- 25 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 486-491
Open AccessUnmasking Propaganda: A Comparative Analysis of Masked and Causal Language Models
- Open Access
Qwen it detect machine-generated text?
This paper describes the approach of the Unibuc - NLP team in tackling the Coling 2025 GenAI Workshop, Task 1: Binary Multilingual Machine-Generated Text Detection. We explored both masked language models and causal models. For Subtask A, our best model achieved first-place out of 36 teams when looking at F1 Micro (Auxiliary Score) of 0.8333, and second-place when looking at F1 Macro (Main Score) of 0.8301
Open AccessQwen it detect machine-generated text?
This paper describes the approach of the Unibuc - NLP team in tackling the Coling 2025 GenAI Workshop, Task 1: Binary Multilingual Machine-Generated Text Detection. We explored both masked language models and causal models. For Subtask A, our best model achieved first-place out of 36 teams when looking at F1 Micro (Auxiliary Score) of 0.8333, and second-place when looking at F1 Macro (Main Score) of 0.8301
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 15309-15323
Friend or Foe? A Computational Investigation of Semantic False Friends across Romance Languages
Ana Sabina Uban, Liviu P Dinu, Ioan-Bogdan Iordache, Simona Georgescu, Claudia Vlad. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 15309-15323
Open AccessFriend or Foe? A Computational Investigation of Semantic False Friends across Romance Languages
Ana Sabina Uban, Liviu P Dinu, Ioan-Bogdan Iordache, Simona Georgescu, Claudia Vlad. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 780-793
Assessing Critical Thinking Components in Romanian Secondary School Textbooks: A Data Mining Approach to the ROTEX Corpus
Madalina Chitez, Liviu Dinu, Marius Micluta-Campeanu, Ana-Maria Bucur, Roxana Rogobete. Proceedings of the 20th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2025). 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 780-793
Open AccessAssessing Critical Thinking Components in Romanian Secondary School Textbooks: A Data Mining Approach to the ROTEX Corpus
Madalina Chitez, Liviu Dinu, Marius Micluta-Campeanu, Ana-Maria Bucur, Roxana Rogobete. Proceedings of the 20th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2025). 2025.
- Open Access
Transformer based neural networks for emotion recognition in conversations
This paper outlines the approach of the ISDS-NLP team in the SemEval 2024 Task 10: Emotion Discovery and Reasoning its Flip in Conversation (EDiReF). For Subtask 1 we obtained a weighted F1 score of 0.43 and placed 12 in the leaderboard. We investigate two distinct approaches: Masked Language Modeling (MLM) and Causal Language Modeling (CLM). For MLM, we employ pre-trained BERT-like models in a multilingual setting, fine-tuning them with a classifier to predict emotions. Experiments with varying input lengths, classifier architectures, and fine-tuning strategies demonstrate the effectiveness of this approach. Additionally, we utilize Mistral 7B Instruct V0.2, a state-of-the-art model, applying zero-shot and few-shot prompting techniques. Our findings indicate that while Mistral shows promise, MLMs currently outperform them in sentence-level emotion classification.
Open AccessTransformer based neural networks for emotion recognition in conversations
This paper outlines the approach of the ISDS-NLP team in the SemEval 2024 Task 10: Emotion Discovery and Reasoning its Flip in Conversation (EDiReF). For Subtask 1 we obtained a weighted F1 score of 0.43 and placed 12 in the leaderboard. We investigate two distinct approaches: Masked Language Modeling (MLM) and Causal Language Modeling (CLM). For MLM, we employ pre-trained BERT-like models in a multilingual setting, fine-tuning them with a classifier to predict emotions. Experiments with varying input lengths, classifier architectures, and fine-tuning strategies demonstrate the effectiveness of this approach. Additionally, we utilize Mistral 7B Instruct V0.2, a state-of-the-art model, applying zero-shot and few-shot prompting techniques. Our findings indicate that while Mistral shows promise, MLMs currently outperform them in sentence-level emotion classification.
- Open Access
Designing NLP Systems That Adapt to Diverse Worldviews
Natural Language Inference (NLI) is foundational for evaluating language understanding in AI. However, progress has plateaued, with models failing on ambiguous examples and exhibiting poor generalization. We argue that this stems from disregarding the subjective nature of meaning, which is intrinsically tied to an individual's \textit{weltanschauung} (which roughly translates to worldview). Existing NLP datasets often obscure this by aggregating labels or filtering out disagreement. We propose a perspectivist approach: building datasets that capture annotator demographics, values, and justifications for their labels. Such datasets would explicitly model diverse worldviews. Our initial experiments with a subset of the SBIC dataset demonstrate that even limited annotator metadata can improve model performance.
Open AccessDesigning NLP Systems That Adapt to Diverse Worldviews
Natural Language Inference (NLI) is foundational for evaluating language understanding in AI. However, progress has plateaued, with models failing on ambiguous examples and exhibiting poor generalization. We argue that this stems from disregarding the subjective nature of meaning, which is intrinsically tied to an individual's \textit{weltanschauung} (which roughly translates to worldview). Existing NLP datasets often obscure this by aggregating labels or filtering out disagreement. We propose a perspectivist approach: building datasets that capture annotator demographics, values, and justifications for their labels. Such datasets would explicitly model diverse worldviews. Our initial experiments with a subset of the SBIC dataset demonstrate that even limited annotator metadata can improve model performance.
- Open Access
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 6314-6326
Verba volant, scripta volant? Don’t worry! There are computational solutions for protoword reconstruction
Liviu P Dinu, Ana Sabina Uban, Alina Maria Cristea, Ioan-Bogdan Iordache, Teodor-George Marchitan, Simona Georgescu, Laurentiu Zoicas. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 6314-6326
Open AccessVerba volant, scripta volant? Don’t worry! There are computational solutions for protoword reconstruction
Liviu P Dinu, Ana Sabina Uban, Alina Maria Cristea, Ioan-Bogdan Iordache, Teodor-George Marchitan, Simona Georgescu, Laurentiu Zoicas. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.