TALN
@taln_ls2n
Followers
68
Following
68
Media
21
Statuses
83
TALN research team - @LaboLS2N | @NantesUniv - @IUTNantes | #NLP · #AI · #CL | https://t.co/WCMXbOFl9P
Nantes, France
Joined November 2021
▶️ Hiring for a 15-month postdoctoral contract with Lexhnology ANR-22-CE38-0004, an exciting interdisciplinary project with AI, NPL, linguistics, and language learning. Bonus: great job location and colleagues
2
2
2
#RésultatScientifique 🔎 | @ArthurAmalvy, @r_dufour et Vincent Labatut montrent que le découpage de longs textes par des modèles de langue peut nuire à leurs performances et y proposent une solution ➡️ https://t.co/H09ObbKa7c 🤝 @CNRS_dr17 @CentraleNantes @LaboLS2N @LIAvignon
0
5
3
Reading Group at @taln_ls2n with @MaelHoubre who just came back from @jouhouken NII. Keyphrase generation rocks !
0
2
2
@CNRSinformatics @LaboLS2N @taln_ls2n @LabrakYanis 🖥️ Demo of BioMistral is now available thanks to the support of @huggingface 🤗 https://t.co/Us6bzVyDCZ
0
3
10
📖 Improving Text Readability through Segmentation into Rheses 🏢 LS2N x MOBiDYS 🖋 Antoine Jamelot, Solen Quiniou, Sophie Hamon (7/7)
0
0
0
💉 How Important Is Tokenization in French Medical Masked Language Models? 🏢 LS2N x LIA x ZENIDOC 🖋 Yanis Labrak, Adrien Bazoge, Béatrice Daille, Mickaël Rouvier, Richard Dufour 🔗 https://t.co/WfnU6bbldT (6/7)
hal.science
Subword tokenization has become the prevailing standard in the field of natural language processing (NLP) over recent years, primarily due to the widespread utilization of pre-trained language...
1
0
0
🩺 A Zero-shot and Few-shot Study of Instruction-Finetuned Large Language Models Applied to Clinical and Biomedical Tasks 🏢 LS2N x LIA x ZENIDOC 🖋 Yanis Labrak, Mickaël Rouvier, Richard Dufour 🔗 https://t.co/RyHFs8F5pI (5/7)
hal.science
The recent emergence of Large Language Models (LLMs) has enabled significant advances in the field of Natural Language Processing (NLP). While these new models have demonstrated superior performance...
1
0
0
🧪 DrBenchmark: A Large Language Understanding Evaluation Benchmark for French Biomedical Domain 🏢 LS2N x LIA x CHU de Nantes x ZENIDOC x STL 🖋 Yanis Labrak @LabrakYanis, Adrien Bazoge, et al. 🔗 https://t.co/AOAviZzKCy (4/7)
hal.science
The biomedical domain has sparked a significant interest in the field of Natural Language Processing (NLP), which has seen substantial advancements with pre-trained language models (PLMs). However,...
1
0
0
📝 CASIMIR: A Corpus of Scientific Articles enhanced with Multiple Author-Integrated Revisions 🏢 LS2N x NII Leane Jourdan @LeaneJourdan , Florian Boudin, Nicolas Hernandez @nicolarnandz, Richard Dufour @r_dufour 🔗 https://t.co/I0E96g84jj (3/7)
hal.science
Writing a scientific article is a challenging task as it is a highly codified and specific genre, consequently proficiency in written communication is essential for effectively conveying research...
1
0
0
We proposed: a corpus of revised scientific articles 📝, a benchmark for the biomedical LM evaluation🧪, an LLM study in zero- and few-shot settings in the medical domain 🩺, a new word tokenization approach for medical words 💉, a method for improving text readability 📖 (2/7)
1
0
0
🎉 Happy to share that the TALN/NLP research group of the LS2N (Nantes University) has 5 accepted papers ✅ at the @LrecColing 2024 conference! 📚✨ (1/7)
1
1
2
⚠️ [Appel à Participation] Campagne évaluation DEFT 2024 ⚠️ 📝Tâche : Réponse automatique à des QCM issus d'annales d'examens de pharmacie 🌐 Plus d'infos : https://t.co/bAvXTfe4oF 🚀N'hésitez pas à participer !
0
4
5
BioMistral is a new 7B foundation model for medical domains, based on Mistral and further trained PubMed Central. - top open-source medical Large Language Model (LLM) in its weight class - Apache License - includes base models, fine tunes, and quantized versions.
20
222
1K
🚀 🏥 Very proud to announce BioMistral, a collection of open-source pre-trained LLMs for the medical domain 📰Arxiv: https://t.co/Zarwhb807S 🏥 BioMistral 7B model: https://t.co/h00IOmXC66 More info: https://t.co/eiFj4bk8b5
@CNRSinformatics @LaboLS2N @taln_ls2n @LabrakYanis
3
20
52
BioMistral 7B just arrived! A Multilingual Medical LLM for the medical domain. BioMistral's multilingual potential and robustness across diverse linguistic contexts, with a focus on truthfulness and calibration in multilingual environments. The 7B model surpasses all competitive
🚀 🏥 Very proud to announce BioMistral, a collection of open-source pre-trained LLMs for the medical domain 📰Arxiv: https://t.co/Zarwhb807S 🏥 BioMistral 7B model: https://t.co/h00IOmXC66 More info: https://t.co/eiFj4bk8b5
@CNRSinformatics @LaboLS2N @taln_ls2n @LabrakYanis
4
28
78
A Survey of Pre-trained Language Models for Processing Scientific Text https://t.co/6DTsCWe39g Xanh Ho, Anh Khoa Duong Nguyen, An Tuan Dao, Junfeng Jiang, Yuki Chida, Kaito Sugimoto, Huy Quoc To, Florian Boudin, Akiko Aizawa
0
2
2
Journée d’études : ce que l’Intelligence Artificielle change à l’Université organisée par les Bibliothèques Universitaires de Nantes et @chaire_RELIA UNESCO
bu.univ-nantes.fr
Poursuite de la journée sur place : Merci à tous les intervenants de la table ronde et animateurs d'ateliers : IA et recherche Doc, IA et bibliométrie, IA et REL, IA et corpus biblio
0
4
4
Fruitful and fun plenary meeting of the Lexhnology consortium in Nancy! We've meet great strides in modelizing SCOTUS case law, collecting user needs and use cases. We'll be doing the full annotation campaign and gearing up for experiments with users this in 2024.
2
4
6
Voici HATS !🎩 Grâce aux 143 personnes qui ont participé à mon expérience, j'ai pu obtenir 7150 annotations et construire le data set HATS que nous diffusons librement afin que la communauté puisse évaluer les métriques de reconnaissance de la Parole🎙️
3
2
7
Anas Belfathi presenting "Harnessing GPT-3.5-turbo for Rhetorical Role Prediction in Legal Cases" #JURIX2023 @maaslawtech #longpaper. Proceeding available on https://t.co/eqPxOUQa66
1
4
2