10 papers
AT-ViT: Area-Targeted Multi-View Vision Transformer with Cross-Attention and Multi-Scale Patching for Plant Trait Recognition in Herbarium Images
Amani Sedrat, Takieddine Chehhat, Youcef Sklab +6
Automated plant traits recognition from herbarium images is essential for plant sciences, yet remains challenging because background elements (e.g., textual labels, mounting artifa…
An Agentic Framework Using Rules and LLMs for Embedding and Annotating Descriptive Document Layouts: A Plant Science Use Case
Nicolas Turenne, Youcef Sklab, Eric Chenin +1
Background: Recent advances in information retrieval (IR) leverage both dense and sparse representations, large language models (LLMs), and specialized retrieval models to improve…
Are Tabular Foundation Models Robust to Realistic Query Distribution Shifts in Microbiome Data?
Giulia Perciballi, Ahmad Fall, Federica Granese +2
Tabular foundation models (TFMs) achieve strong performance on microbiome abundance data, yet their robustness under realistic distribution shift remains poorly characterized. We i…
MetagenBERT: a Transformer-based Architecture using Foundational genomic Large Language Models for novel Metagenome Representation
Gaspar Roy, Eugeni Belda, Baptiste Hennecart +3
Metagenomic disease prediction commonly relies on species abundance tables derived from large, incomplete reference catalogs, constraining resolution and discarding valuable inform…
A text-to-tabular approach to generate synthetic patient data using LLMs
Margaux Tornqvist, Jean-Daniel Zucker, Tristan Fauvel +3
Access to large-scale high-quality healthcare databases is key to accelerate medical research and make insightful discoveries about diseases. However, access to such data is often…
SIM-Net: A Multimodal Fusion Network Using Inferred 3D Object Shape Point Clouds from RGB Images for 2D Classification
Youcef Sklab, Hanane Ariouat, Eric Chenin +2
We introduce the Shape-Image Multimodal Network (SIM-Net), a novel 2D image classification architecture that integrates 3D point cloud representations inferred directly from RGB im…