Roberto Interdonato
banner
interdonatos.bsky.social
Roberto Interdonato
@interdonatos.bsky.social
Senior Researcher @ CIRAD, UMR TETIS (Montpellier) and INRIA team-project EVERGREEN (Earth observation and machine learning for agro-environmental challenges). AI/complex networks/food security/agriculture. But also good food/music/literature/photography.
📢 Call for Papers Now Open! 📢

Our special issue on Neurocomputing ( @elsevierconnect.bsky.social ) "Advanced Machine Learning Techniques for Earth Observation Data Analysis: From Multi-Modal Sensing to Intelligent Environmental Monitoring" is now open!

Deadline: March 2nd, 2026

lnkd.in/dRw2Z3sv
November 5, 2025 at 7:52 AM
Reposted by Roberto Interdonato
Our latest paper has just been published. We explore geographical biases in language models and their implications (with a focus on crisis monitoring).
@interdonatos.bsky.social , @matroche.bsky.social , M.Teisseire & S.Valentin.

Code: github.com/tetis-nlp/ge...

link.springer.com/article/10.1...
Evaluation of geographical distortions in language models - Machine Learning
Geographic bias in language models (LMs) is an underexplored dimension of model fairness, despite growing attention being given to other social biases. We investigate whether LMs provide equally accur...
link.springer.com
October 28, 2025 at 1:21 PM
📢 Call for Papers

Excited to share that submissions are now open for the ECML PKDD 2026 Journal Track! 🚀

2 Submission Deadlines:
1st : October 30, 2025
2nd : January 15, 2026

🔗Full details: ecmlpkdd.org/2025/submiss...

#MachineLearning #KnowledgeDiscovery #ECMLPKDD @ecmlpkdd.org #AI
1st Submission Deadline
ecmlpkdd.org
October 1, 2025 at 8:20 AM
Reposted by Roberto Interdonato
We are looking for an NLP postdoc/engineer to work on adding language capabilities to our Earth observation sensor-agnostic models (Atomizer, to be presented at BMVC25).

Details here: jobs.inria.fr/public/class...

Atomizer: arxiv.org/pdf/2506.13542
GEO-ReSeT project: anr.fr/Projet-ANR-2...
September 8, 2025 at 12:30 PM
Reposted by Roberto Interdonato
We show that our multi-sensor approach is more robust in average than recent methods from the EO literature in three classification tasks, namely cropland classification, crop-type classification, and tree-species classification.

@interdonatos.bsky.social
May 13, 2025 at 11:37 AM
Not sure why I'm here but here's a nice photo taken at the Jardin des Plantes in Montpellier with my Olympus OM-D E-M10 Mark II + Olympus 45mm 1.8
December 17, 2024 at 1:19 PM