Sahar Ghannay
Présentation
Team: ILES Email: sahar.ghannay@limsi.fr Website: https://saharghannay.github.io Sahar Ghannay is an associate professor at Université Paris-Saclay, in the CNRS, LISN research center, since September 2018. She received a PhD in Computer Science from Le Mans University on Septembre 2017. Her thesis work is part of the ANR VERA (AdVanced ERror Analysis for speech recognition) project. During her PhD, she spent a few months as @ visiting researcher at Apple within the Siri Speech team. As a postdoctoral researcher at LIUM, she worked on neural end-to-end systems for the detection of named entities, speech understanding, as part of the Chist-Era M2CR (Multimodal Multilingual Continuous Representation for Human Language Understanding) project. Her research interests are representation learning and semantic information extraction from spoken and written language.
More recently, her work has extended to multimodal representations, particularly through visual question answering and the development of multimodal corpora.Associate professort at LIMSI, CNRS, Université Paris-Saclay
Short Bio
CV
Education
Work Experience
Publications
Publications
|
|
Evaluating the Confidentiality of Synthetic Clinical Texts Generated by Language Models23rd International Conference on Artificial Intelligence in Medicine (AIME), Jun 2025, Pavie, Italy. pp.130-139, ⟨10.1007/978-3-031-95838-0_13⟩ |
|
|
Entity-aware cross-modal pretraining for Knowledge-Based Visual Question AnsweringEuropean Conference on Information Retrieval (ECIR), Apr 2025, Lucca, Italy. pp.391-400, ⟨10.1007/978-3-031-88714-7_38⟩ |
|
|
Génération augmentée de récupération multi-niveau pour répondre à des questions visuelles20e Conférence en Recherche d’Information et Applications (CORIA) 32ème Conférence sur le Traitement Automatique des Langues Naturelles (TALN) 27ème Rencontre des Étudiants Chercheurs en Informatique pour le Traitement Automatique des Langues (RECITAL) Les 18e Rencontres Jeunes Chercheurs en RI (RJCRI), Jul 2025, Marseille, France. pp.128-130 |
|
|
XAI for Gender Representation in Media Analysis2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2025), IEEE Signal Processing Society, Apr 2025, Hyderabad, India. pp.1-5, ⟨10.1109/ICASSP49660.2025.10888945⟩ |
|
|
Détection des contaminations de LLM par extraction de données : une revue de littérature pratique20e Conférence en Recherche d’Information et Applications (CORIA) 32ème Conférence sur le Traitement Automatique des Langues Naturelles (TALN) 27ème Rencontre des Étudiants Chercheurs en Informatique pour le Traitement Automatique des Langues (RECITAL) Les 18e Rencontres Jeunes Chercheurs en RI (RJCRI), Jul 2025, Marseille, France. pp.233-251 |
|
|
Leveraging Information Retrieval to Enhance Spoken Language Understanding Prompts in Few-Shot LearningInterspeech 2025, Aug 2025, Rotterdam, Netherlands. ⟨10.21437/Interspeech.2025-175⟩ |
|
|
A dual task learning approach to fine-tune a multilingual semantic speech encoder for Spoken Language UnderstandingInterspeech 2024, Sep 2024, Kos, Greece |
|
|
New Semantic Task for the French Spoken Language Understanding MEDIA BenchmarkThe 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), May 2024, Torino, Italy. pp.12227-12246 |
|
|
Les petits modèles sont bons : une étude empirique de classification dans un contexte zero-shot31ème Conférence sur le Traitement Automatique des Langues Naturelles (TALN 2024), Jul 2024, Toulouse, France. pp.113-129 |
|
|
Nouvelle tâche sémantique pour le corpus de compréhension de parole en français MEDIA35èmes Journées d'Études sur la Parole (JEP 2024), Jul 2024, Toulouse, France. pp.470-480 |
|
|
Small Language Models are Good Too: An Empirical Study of Zero-Shot ClassificationThe 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), May 2024, TURIN, Italy. pp.14923-14936 |
|
|
Multi-Level Information Retrieval Augmented Generation for Knowledge-based Visual Question AnsweringThe 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP 2024), Nov 2024, Miami, United States. pp.16499-16513, ⟨10.18653/v1/2024.emnlp-main.922⟩ |
|
|
mALBERT: Is a Compact Multilingual BERT Model Still Worth It?The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), May 2024, Torino, Italy. pp.11023-11029 |
|
|
Utiliser l'explicabilité des modèles pour mettre en évidence les expressions genrées dans la parole31ème Conférence sur le Traitement Automatique des Langues Naturelles (TALN 2024), Jul 2024, Toulouse, France. pp.695-707 |
|
|
Projet Gender Equality Monitor (GEM)CORIA-TALN 2023 30e Conférence sur le Traitement Automatique des Langues Naturelles (TALN), Jun 2023, Paris, France. pp.21-21 |
|
|
Explicit knowledge integration for knowledge-aware visual question answering about named entitiesICRM'23 - ACM International Conference on Multimedia Retrieval, ACM, Jun 2023, Thessalonique, Greece. pp.29-38, ⟨10.1145/3591106.3592227⟩ |
Specialized Semantic Enrichment of Speech Representations2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW), Jun 2023, Rhodes Island, Greece. pp.1-5, ⟨10.1109/ICASSPW59220.2023.10193452⟩ |
|
|
|
Continual self-supervised domain adaptation for end-to-end speaker diarizationIEEE Spoken Language Technology Workshop (SLT 2022), IEEE Speech and Language Processing Technical Committee, Jan 2023, Doha, Qatar. pp.626 - 632, ⟨10.1109/SLT54892.2023.10023195⟩ |
|
|
Analyzing BERT Cross-lingual Transfer Capabilities in Continual Sequence LabelingFirst Workshop on Performance and Interpretability Evaluations of Multimodal, Multipurpose, Massive-Scale Models, Oct 2022, virtual, South Korea |
|
|
The Spoken Language Understanding MEDIA Benchmark Dataset in the Era of Deep Learning: data updates, training and evaluation toolsLREC 2022 - 13th Language Resources and Evaluation Conference, Jun 2022, Marseille, France. pp.1595‑1602 |
|
|
Impact Analysis of the Use of Speech and Language Models Pretrained by Self-Supersivion for Spoken Language UnderstandingLREC 2022 - 13th Language Resources and Evaluation Conference, Jun 2022, Marseille, France. pp.2949‑2956 |
|
|
Étude comparative de modèles Transformers en compréhension de la parole en français34e Journées d'Etudes sur la Parole (JEP2022), Jun 2022, Île de Noirmoutier, France. pp.721-730, ⟨10.21437/JEP.2022-76⟩ |
|
|
Benchmarking Transformers-based models on French Spoken Language Understanding tasksINTERSPEECH 2022, Sep 2022, Incheon, South Korea. pp.1238-1242, ⟨10.21437/Interspeech.2022-385⟩ |
|
|
Le benchmark MEDIA revisité : données, outils et évaluation dans un contexte d'apprentissage profondXXXIVe Journées d'Études sur la Parole -- JEP 2022, Jun 2022, île de Noirmoutier, France. pp.481-490, ⟨10.21437/JEP.2022-51⟩ |
|
|
Evaluating the carbon footprint of NLP methods: a survey and analysis of existing toolsEMNLP, Workshop SustaiNLP, Nov 2021, Punta Cana, Dominican Republic. ⟨10.18653/v1/2021.sustainlp-1.2⟩ |
|
|
Overlap-Aware Low-Latency Online Speaker Diarization Based on End-To-End Local SegmentationIEEE Automatic Speech Recognition and Unserstanding Workshop, Dec 2021, Cartagena, Colombia. pp.1139-1146, ⟨10.1109/ASRU51503.2021.9688044⟩ |
|
|
Where are we in semantic concept extraction for Spoken Language Understanding? ⋆SPECOM 2021 23rd International Conference on Speech and Computer, Sep 2021, Saint Petersburg, Russia |
|
|
Experiments from LIMSI at the French Named Entity Recognition Coarse-grained taskConference and Labs of the Evaluation Forum, Sep 2020, Thessaloniki, Greece |
|
|
A Metric Learning Approach to Misogyny CategorizationWorkshop on Representation Learning for NLP, Jul 2020, Online, France. pp.89-94, ⟨10.18653/v1/2020.repl4nlp-1.12⟩ |
Error analysis applied to end-to end spoken language understanding45th International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2020), May 2020, Barcelona, Spain. pp.8514-8518, ⟨10.1109/ICASSP40776.2020.9054455⟩ |
|
|
|
What is best for Spoken Language Understanding: Small but Task-dependant Embeddings or Huge but Out-of-domain Embeddings?45th International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2020), May 2020, Barcelona, Spain. pp.8114-8118, ⟨10.1109/ICASSP40776.2020.9053278⟩ |
|
|
A Cooking Knowledge Graph and Benchmark for Question Answering Evaluation in Lifelong Learning ScenariosInternational Conference on Applications of Natural Language to Information Systems, Elisabeth Métais and Farid Meziane and Helmut Horacek and Philipp Cimiano, Jun 2020, Saarbrücken, Germany |
|
|
Neural Networks approaches focused on French Spoken Language Understanding: application to the MEDIA Evaluation TaskIn Proceedings of The 28th International Conference on Computational Linguistics (COLING’2020), 2020, Dec 2020, Barcelona (online), Spain. pp.2722-2727, ⟨10.18653/v1/2020.coling-main.245⟩ |
|
|
A Comparison of Metric Learning Loss Functions for End-To-End Speaker VerificationInternational Conference on Statistical Language and Speech Processing, Oct 2020, Cardiff, United Kingdom. pp.137-148, ⟨10.1007/978-3-030-59430-5_11⟩ |
|
|
Lifelong learning and task-oriented dialogue system: what does it mean?International Workshop on Spoken Dialogue Systems Technology, Apr 2019, Siracusa, Italy |
|
|
Simulating ASR errors for training SLU systemsLREC 2018, May 2018, Miyazaki, Japan |
Task Specific Sentence Embeddings for ASR Error DetectionInterspeech 2018, Sep 2018, Hyderabad, India. ⟨10.21437/Interspeech.2018-2211⟩ |
|
|
|
Simulation d'erreurs de reconnaissance automatique dans un cadre de compréhension de la paroleXXXIIe Journées d'Etudes sur la Parole (JEP 2018), Jun 2018, Aix-en-Provence, France |
Représentations de phrases dans un espace continu spécifiques à la tâche de détection d'erreursXXXIIe Journées d'Etudes sur la Parole (JEP 2018), Jun 2018, Aix-en-Provence, France |
|
|
|
End-to-end named entity and semantic concept extraction from speechIEEE Spoken Language Technology Workshop, Dec 2018, Athens, Greece. ⟨10.1109/slt.2018.8639513⟩ |
|
|
ASR error management for improving spoken language understandingInterspeech 2017, Aug 2017, Stockholm, Sweden |
|
|
Enriching confusion networks for post-processingStatistical Language and Speech Processing 2017, Oct 2017, Le Mans, France |
Recent improvements on error detection for automatic speech recognition1st International Workshop on Multimodal Media Data Analytics (MMDA 2016), in Conjunction with the 22nd European Conference on Artificial Intelligence, 2016, The Hague The, Netherlands |
|
Utilisation des représentations continues des mots et des paramètres prosodiques pour la détection des erreurs dans les transcriptions automatiques de la paroleJournée d'Études sur la Parole (2016), Jul 2016, Paris, France. pp.723-731 |
|
Word embedding evaluation and combination10th edition of the Language Resources and Evaluation Conference (LREC 2016), 2016, Portorož, Slovenia |
|
Evaluation of acoustic word embeddingsRepEval@ACL 2016: The 1st Workshop on Evaluating Vector-Space Representations for NLP, 2016, Berlin, Germany |
|
Acoustic word embeddings for ASR error detectionInterspeech 2016, 2016, San Francisco (CA, USA), Unknown Region |
|
Combining continous word representation and prosodic features for ASR error prediction3rd International Conference on Statistical Language and Speech Processing (SLSP 2015), 2015, Budapest, Hungary |
|
Which ASR errors are hard to detect?Workshop Errors by Humans and Machines in multimedia, multimodal and multilingual data processing (ERRARE 2015), 2015, Sinaia, Romania |
|
Word embeddings combination and neural networks for robustness in ASR error detection2015 European Signal Processing Conference (EUSIPCO 2015), 2015, Nice, France |
|
Using Hypothesis Selection Based Features for Confusion Network MT System CombinationThird Workshop on Hybrid Approaches to Translation (HyTra), EACL 2014, 2014, Gothenburg, Sweden |
|
|
AtomicEval: Evaluation Framework for Atomic Proposition Autonomy with French Propositioner10th Junior Conference on Data Sciences and Engineering, Sep 2025, Gif-sur-Yvette, France. |
|
|
Has AlphaFold3 achieved success for RNAs?2025 |
|
|
LIMSI_UPV at SemEval-2020 Task 9: Recurrent Convolutional Neural Network for Code-mixed Sentiment Analysis2021 |
|
|
Semantic enrichment towards efficient speech representationsLISN. 2023 |
|
|
Étude sur les représentations continues de mots appliquées à la détection automatique des erreurs de reconnaissance de la paroleInformatique et langage [cs.CL]. Université du Maine, 2017. Français. ⟨NNT : 2017LEMA1019⟩ |