biblissim-ia-2024 | Intelligence artificielle et la reconnaissance de formes et d'écritures manuscrites Journées annuelles du cluster 3 de l'EquipEx Biblissima+

9:15 - 10:00 (45min)

Accueil - Petit déjeuner

Foyer

10:00 - 10:30 (30min)

Ansund: Using Machine Learning to Develop a New, Exhaustive, Open Access Corpus of Old English

Auditorium 150

M. Faulkner , E. Magnanti

https://biblissim-ia-2024.sciencesconf.org/data/program/Faulkner_and_Magnanti_Biblissima24.pdf

10:30 - 11:15 (45min)

Keynote speech : Signs, symbols and secrets: HTR faced with ciphered texts and music scores

Auditorium 150

A. Fornes

11:15 - 11:30 (15min)

Pause café

Foyer

11:30 - 12:00 (30min)

Deep Digital oriental: A state of the art for Sanskrit oriental handwritten text recognition

Auditorium 150

S. Thottempudi

https://biblissim-ia-2024.sciencesconf.org/data/program/Thottempudi_Biblissima_2025.pdf

12:00 - 12:30 (30min)

Intégration de modèles Seq2Seq pour la correction et la normalisation de textes issus de l'HTR

Auditorium 150

S. Yatsyk

https://biblissim-ia-2024.sciencesconf.org/data/program/YatsykBiblissima_2025.pdf

12:30 - 13:00 (30min)

Entraînement from scratch ou fine-tuning ? Modalités et enjeux d'un choix d'un projet HTR sur les registres de plaidoiries du Parlement de Paris (fin XIVe-XVe siècle)

Auditorium 150

P. Spychala

https://biblissim-ia-2024.sciencesconf.org/data/program/Spychala_Biblissima_2025.pdf

13:00 - 14:00 (1h)

Déjeuner

Foyer

13:00 - 14:00 (1h)

Session poster

Foyer

14:00 - 14:30 (30min)

How Do Large Language Models (LLMs) Impact Document Image Analysis? An Open Discussion.

Auditorium 150

A. Scius-Bertrand, L. Voegtlin, N. Wegmann, A. Fakhari, A. Fischer

https://biblissim-ia-2024.sciencesconf.org/data/program/LLM_DocAnalysis_submit_1.pdf

14:30 - 15:15 (45min)

TR and LLM feedback for automatic transcription in Divergent Use Cases

Auditorium 150

H. Miller, M. Lavee, N. Bontemps

15:15 - 15:45 (30min)

Integrating Vision Transformers and Large Language Models for Enhanced Handwriting Text Recognition and Named Entity Identification in Documentary Manuscripts

Auditorium 150

S. Aguilar Torres

https://biblissim-ia-2024.sciencesconf.org/data/program/integrating_VIT_LLM_HTR_1.pdf

15:45 - 16:00 (15min)

Pause café

Foyer

16:00 - 16:30 (30min)

« Un simple OCR ? » : outils, algorithmes et méthodes de reconnaissance de document en humanité numérique.

Auditorium 150

C. Kermorvant , Y. Schneider, M. Blanco

https://biblissim-ia-2024.sciencesconf.org/data/program/teklia_biblissima24.pdf

16:30 - 17:00 (30min)

Retour(s) d'expérience(s) d'indexation de registres d'emprunteurs de bibliothèques parisiennes: chaîne de production et de diffusion des archives de PRET19

Auditorium 150

V. Rebolledo-Dhuin, C. Kermorvant

https://biblissim-ia-2024.sciencesconf.org/data/program/Viera_Biblissima24.pdf

17:00 - 17:45 (45min)

Keynote speech : Information extraction from noisy text output

Auditorium 150

A. Doucet

Programme