Romaric Besançon
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Romaric Besançon has a total of 67 co-authors. Here are the top co-authors by number of shared publications:
- Olivier Ferret
Université Paris-Saclay, CEA, List, Palaiseau, France
11 shared publications Olivier Ferret
10 shared publicationsBrigitte Grau
Université Paris-Saclay, CNRS, LIMSI, ENSIIE
5 shared publications- Hervé Le Borgne
Université Paris-Saclay, CEA, List
4 shared publications - Aboubacar Tuo
Université Paris-Saclay, CEA, List
3 shared publications Christian Fluhr
CEA-LIST
3 shared publications
Latest Publications
- 24 August 2026
- Published by Springer Nature
- p. 575-591
GLiDRE: Generalist Lightweight Model for Document-Level Relation Extraction
- 24 August 2026
- Published by Springer Nature
- p. 575-591
GLiDRE: Generalist Lightweight Model for Document-Level Relation Extraction
- Open Access
Explaining How Visual, Textual and Multimodal Encoders Share Concepts
Sparse autoencoders (SAEs) have emerged as a powerful technique for extracting human-interpretable features from neural networks activations. Previous works compared different models based on SAE-derived features but those comparisons have been restricted to models within the same modality. We propose a novel indicator allowing quantitative comparison of models across SAE features, and use it to conduct a comparative study of visual, textual and multimodal encoders. We also propose to quantify the Comparative Sharedness of individual features between different classes of models. With these two new tools, we conduct several studies on 21 encoders of the three types, with two significantly different sizes, and considering generalist and domain specific datasets. The results allow to revisit previous studies at the light of encoders trained in a multimodal context and to quantify to which extent all these models share some representations or features. They also suggest that visual features that are specific to VLMs among vision encoders are shared with text encoders, highlighting the impact of text pretraining. The code is available at https://github.com/CEA-LIST/SAEshareConcepts
Open AccessExplaining How Visual, Textual and Multimodal Encoders Share Concepts
Sparse autoencoders (SAEs) have emerged as a powerful technique for extracting human-interpretable features from neural networks activations. Previous works compared different models based on SAE-derived features but those comparisons have been restricted to models within the same modality. We propose a novel indicator allowing quantitative comparison of models across SAE features, and use it to conduct a comparative study of visual, textual and multimodal encoders. We also propose to quantify the Comparative Sharedness of individual features between different classes of models. With these two new tools, we conduct several studies on 21 encoders of the three types, with two significantly different sizes, and considering generalist and domain specific datasets. The results allow to revisit previous studies at the light of encoders trained in a multimodal context and to quantify to which extent all these models share some representations or features. They also suggest that visual features that are specific to VLMs among vision encoders are shared with text encoders, highlighting the impact of text pretraining. The code is available at https://github.com/CEA-LIST/SAEshareConcepts
- 20 September 2024
- Published by Springer Nature in Language Resources and Evaluation
- Vol. 58 (4) , 1463-1480
From LIMA to DeepLIMA: following a new path of interoperability
- 20 September 2024
- Published by Springer Nature in Language Resources and Evaluation
- Vol. 58 (4) , 1463-1480
From LIMA to DeepLIMA: following a new path of interoperability
- Open Access
From LIMA to DeepLIMA: following a new path of interoperability
In this article, we describe the architecture of the LIMA (Libre Multilingual Analyzer) framework and its recent evolution with the addition of new text analysis modules based on deep neural networks. We extended the functionality of LIMA in terms of the number of supported languages while preserving existing configurable architecture and the availability of previously developed rule-based and statistical analysis components. Models were trained for more than 60 languages on the Universal Dependencies 2.5 corpora, WikiNer corpora, and CoNLL-03 dataset. Universal Dependencies allowed us to increase the number of supported languages and to generate models that could be integrated into other platforms. This integration of ubiquitous Deep Learning Natural Language Processing models and the use of standard annotated collections using Universal Dependencies can be viewed as a new path of interoperability, through the normalization of models and data, that are complementary to a more standard technical interoperability, implemented in LIMA through services available in Docker containers on Docker Hub.
Open AccessFrom LIMA to DeepLIMA: following a new path of interoperability
In this article, we describe the architecture of the LIMA (Libre Multilingual Analyzer) framework and its recent evolution with the addition of new text analysis modules based on deep neural networks. We extended the functionality of LIMA in terms of the number of supported languages while preserving existing configurable architecture and the availability of previously developed rule-based and statistical analysis components. Models were trained for more than 60 languages on the Universal Dependencies 2.5 corpora, WikiNer corpora, and CoNLL-03 dataset. Universal Dependencies allowed us to increase the number of supported languages and to generate models that could be integrated into other platforms. This integration of ubiquitous Deep Learning Natural Language Processing models and the use of standard annotated collections using Universal Dependencies can be viewed as a new path of interoperability, through the normalization of models and data, that are complementary to a more standard technical interoperability, implemented in LIMA through services available in Docker containers on Docker Hub.
- Open Access
Automatic Die Studies for Ancient Numismatics
Die studies are fundamental to quantifying ancient monetary production, providing insights into the relationship between coinage, politics, and history. The process requires tedious manual work, which limits the size of the corpora that can be studied. Few works have attempted to automate this task, and none have been properly released and evaluated from a computer vision perspective. We propose a fully automatic approach that introduces several innovations compared to previous methods. We rely on fast and robust local descriptors matching that is set automatically. Second, the core of our proposal is a clustering-based approach that uses an intrinsic metric (that does not need the ground truth labels) to determine its critical hyper-parameters. We validate the approach on two corpora of Greek coins, propose an automatic implementation and evaluation of previous baselines, and show that our approach significantly outperforms them.
Open AccessAutomatic Die Studies for Ancient Numismatics
Die studies are fundamental to quantifying ancient monetary production, providing insights into the relationship between coinage, politics, and history. The process requires tedious manual work, which limits the size of the corpora that can be studied. Few works have attempted to automate this task, and none have been properly released and evaluated from a computer vision perspective. We propose a fully automatic approach that introduces several innovations compared to previous methods. We rely on fast and robust local descriptors matching that is set automatically. Second, the core of our proposal is a clustering-based approach that uses an intrinsic metric (that does not need the ground truth labels) to determine its critical hyper-parameters. We validate the approach on two corpora of Greek coins, propose an automatic implementation and evaluation of previous baselines, and show that our approach significantly outperforms them.
- 16 March 2024
- Published by Springer Nature
- p. 126-142
Probing Pretrained Language Models with Hierarchy Properties
- 16 March 2024
- Published by Springer Nature
- p. 126-142
Probing Pretrained Language Models with Hierarchy Properties
- Open Access
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 146-153
Few-Shot Event Argument Extraction Based on a Meta-Learning Approach
Aboubacar Tuo, Romaric Besançon, Olivier Ferret, Julien Tourille. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 4: Student Research Workshop). 2024.
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 146-153
Open AccessFew-Shot Event Argument Extraction Based on a Meta-Learning Approach
Aboubacar Tuo, Romaric Besançon, Olivier Ferret, Julien Tourille. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 4: Student Research Workshop). 2024.
- 17 March 2023
- Published by Springer Nature
- p. 637-645
Trigger or not Trigger: Dynamic Thresholding for Few Shot Event Detection
- 17 March 2023
- Published by Springer Nature
- p. 637-645
Trigger or not Trigger: Dynamic Thresholding for Few Shot Event Detection
- 13 June 2022
- Published by Springer Nature
- p. 291-298
Better Exploiting BERT for Few-Shot Event Detection
- 13 June 2022
- Published by Springer Nature
- p. 291-298
Better Exploiting BERT for Few-Shot Event Detection