As summer approaches, I am pleased to share my reading list for the season. This compilation is designed to encompass papers that are closely aligned with my research interests, while also including a diverse range of topics (curiosity). If you have any papers you’ve enjoyed—old or new—please share them with me on Twitter or via email. Happy reading!
-
Self-Supervised Learning with Lie Symmetries for Partial Differential Equations, Mialon et al Link
-
EM Distillation for One-step Diffusion Models, Xie et al. Link
-
OmniSat: Self-Supervised Modality Fusion for Earth Observation, Astruc et al. Link
-
Cross-sensor self-supervised training and alignment for remote sensing, Marsocci et al. Link
-
Open-Canopy: A Country-Scale Benchmark for Canopy Height Estimation at Very High Resolution, Fogel et al. Link
-
Gromov-Wasserstein Averaging of Kernel and Distance Matrices, Peyre et al. Link
-
Context-Guided Diffusion for Out-of-Distribution Molecular and Protein Design, Klarner et al. Link
-
Neural Optimal Transport with Lagrangian Costs, Liang-Chieh Chen et al. Link
-
Space-Time Continuous PDE Forecasting using Equivariant Neural Fields, Knigge et al. Link
-
Foundation Models for Generalist Geospatial Artificial Intelligence, Jakubik et al. Link
-
STONE: Self-Supervised Tonality Estimator, Kong et al. Link
-
The Mathematics of Emergence, Cucker et al. Link
-
Generative Data Assimilation of Sparse Weather Station Observations at Kilometer Scales, Manshausen et al. Link
-
Flow Map Matching, Boffi et al. Link
-
Learning Diffusion at Lightspeed, Terpin et al. Link
-
3D Gaussian Splatting for Real-Time Radiance Field Rendering, Kerbl et al. Link
-
Simplified and Generalized Masked Diffusion for Discrete Data, Shi et al. Link
-
Chameleon: Mixed-Modal Early-Fusion Foundation Models, Chameleon Team. Link
-
Estimating Canopy Height at Scale, Pauls et al. Link
-
Generating and Imputing Tabular Data via Diffusion and Flow-based Gradient-Boosted Trees, Jolicoeur-Martineau et al. Link
-
Neural Operators with Localized Integral and Differential Kernels, Liu-Schiaffini et al. Link
-
D-Flow: Differentiating through Flows for Controlled Generation, Ben-Hamu et al. Link