Karin Mora
@karinmora.bsky.social
Researcher @uniLeipzig.bsky.social & @rsc4earth.bsky.social
My work: #PHENOPLEXITY
= #nonlinearDynamics + #complexity + #ML
Analyses: plant #phenology & #biodiversity.
Data: #citizenscience, #remoteSensing, #sentinel2, #eddycovariance, #DataCubes, #TLS
My work: #PHENOPLEXITY
= #nonlinearDynamics + #complexity + #ML
Analyses: plant #phenology & #biodiversity.
Data: #citizenscience, #remoteSensing, #sentinel2, #eddycovariance, #DataCubes, #TLS
Lead by our PhD student Julia Peters
With: @miguelmahecha.bsky.social @chaonanji.bsky.social @dmlmont.bsky.social @cmosig.bsky.social @gdkrmr.bsky.social
Thanks for the support Anca Anghelea
#ESA #AIForGood #RemoteSensing
Preprint here: arxiv.org/abs/2511.11706
With: @miguelmahecha.bsky.social @chaonanji.bsky.social @dmlmont.bsky.social @cmosig.bsky.social @gdkrmr.bsky.social
Thanks for the support Anca Anghelea
#ESA #AIForGood #RemoteSensing
Preprint here: arxiv.org/abs/2511.11706
Context-Aware Multimodal Representation Learning for Spatio-Temporally Explicit Environmental modelling
Earth observation (EO) foundation models have emerged as an effective approach to derive latent representations of the Earth system from various remote sensing sensors. These models produce embeddings...
arxiv.org
November 18, 2025 at 1:14 PM
Lead by our PhD student Julia Peters
With: @miguelmahecha.bsky.social @chaonanji.bsky.social @dmlmont.bsky.social @cmosig.bsky.social @gdkrmr.bsky.social
Thanks for the support Anca Anghelea
#ESA #AIForGood #RemoteSensing
Preprint here: arxiv.org/abs/2511.11706
With: @miguelmahecha.bsky.social @chaonanji.bsky.social @dmlmont.bsky.social @cmosig.bsky.social @gdkrmr.bsky.social
Thanks for the support Anca Anghelea
#ESA #AIForGood #RemoteSensing
Preprint here: arxiv.org/abs/2511.11706
Result 2: #Embeddings that capture ecologically meaningful patterns. We demonstrate this by successfully modelling #GrossPrimaryProduction proving our representations retain the temporal fidelity needed for fine-scale ecosystem analysis.
November 18, 2025 at 1:14 PM
Result 2: #Embeddings that capture ecologically meaningful patterns. We demonstrate this by successfully modelling #GrossPrimaryProduction proving our representations retain the temporal fidelity needed for fine-scale ecosystem analysis.
Result 1: By independently modelling each sensor & then fusing them with a #transformer, our approach is not only powerful but also easily extendable to new data sources.
#DeepLearning #GeoAI #SpatioTemporalModelling #representationLearning #FoundationModels
#DeepLearning #GeoAI #SpatioTemporalModelling #representationLearning #FoundationModels
November 18, 2025 at 1:14 PM
Result 1: By independently modelling each sensor & then fusing them with a #transformer, our approach is not only powerful but also easily extendable to new data sources.
#DeepLearning #GeoAI #SpatioTemporalModelling #representationLearning #FoundationModels
#DeepLearning #GeoAI #SpatioTemporalModelling #representationLearning #FoundationModels
Challenge: #EarthObservation foundation models are often fixed at specific scales, missing the fine spatio-temporal details.
Our Solution: New framework for
🛰️Native 10m spatial resolution
🛰️High temporal frequency
🛰️A unified, analysis-ready latent space that preserves coherence across space & time.
Our Solution: New framework for
🛰️Native 10m spatial resolution
🛰️High temporal frequency
🛰️A unified, analysis-ready latent space that preserves coherence across space & time.
November 18, 2025 at 1:14 PM
Challenge: #EarthObservation foundation models are often fixed at specific scales, missing the fine spatio-temporal details.
Our Solution: New framework for
🛰️Native 10m spatial resolution
🛰️High temporal frequency
🛰️A unified, analysis-ready latent space that preserves coherence across space & time.
Our Solution: New framework for
🛰️Native 10m spatial resolution
🛰️High temporal frequency
🛰️A unified, analysis-ready latent space that preserves coherence across space & time.
Reposted by Karin Mora
Read more about the projects: www.erga-biodiversity.eu/post/erga-at...
@rmwaterhouse.bsky.social @ikeririsarri.bsky.social
@rmwaterhouse.bsky.social @ikeririsarri.bsky.social
ERGA at the BioHackathon Europe 2025
Are you interested in bioinformatics and eager to connect with peers worldwide while working on projects with direct applications for the life sciences? Join this year’s BioHackathon Europe between 3–...
www.erga-biodiversity.eu
November 14, 2025 at 12:32 PM
Read more about the projects: www.erga-biodiversity.eu/post/erga-at...
@rmwaterhouse.bsky.social @ikeririsarri.bsky.social
@rmwaterhouse.bsky.social @ikeririsarri.bsky.social
When listening to CDs we are more likely to listen to the full album. And albums tell a story…
November 15, 2025 at 10:10 PM
When listening to CDs we are more likely to listen to the full album. And albums tell a story…
Really great and insightful photos!
November 14, 2025 at 8:37 AM
Really great and insightful photos!
We have to call things by their name. Schrödinger was a documented Paedophile, as was noted by other commenters. Paedophilia is not a dalliance! It is a crime! Being smart is not a get out of jail free card!
eduexe.co.uk/we-need-to-t...
eduexe.co.uk/we-need-to-t...
EduExe blog We need to talk about Schrödinger
We need to talk about Schrödinger
eduexe.co.uk
November 12, 2025 at 9:29 PM
We have to call things by their name. Schrödinger was a documented Paedophile, as was noted by other commenters. Paedophilia is not a dalliance! It is a crime! Being smart is not a get out of jail free card!
eduexe.co.uk/we-need-to-t...
eduexe.co.uk/we-need-to-t...
