Scholar
Artur P. Toshev
Google Scholar ID: Ck8g7NQAAAAJ
Technical University of Munich
geometric deep learning
graph neural networks
atoms
fluid mechanics
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
90
H-index
6
i10-index
6
Publications
11
Co-authors
22
list available
Contact
Email
artur.toshev@tum.de
CV
Open ↗
Twitter
Open ↗
GitHub
Open ↗
LinkedIn
Open ↗
Publications
3 items
Enhancing Diffusion-Based Sampling with Molecular Collective Variables
2025
Cited
0
LaM-SLidE: Latent Space Modeling of Spatial Dynamical Systems via Linked Entities
2025
Cited
0
Retro-Rank-In: A Ranking-Based Approach for Inorganic Materials Synthesis Planning
2025
Cited
0
Resume (English only)
Academic Achievements
Published 'Neural SPH: Improved Neural Modeling of Lagrangian Fluid Dynamics' at ICML 2024.
Published 'LagrangeBench: A Lagrangian Fluid Mechanics Benchmarking Suite' at NeurIPS 2023 (Datasets and Benchmarks Track).
Published 'Accelerating Molecular Graph Neural Networks via Knowledge Distillation' at NeurIPS 2023 / LOG 2023 (oral presentation).
Published work on E(3)-equivariant GNNs for particle-based fluid simulation at GSI 2023 and ICLR 2023 Physics4ML workshop.
Multiple papers accepted at top-tier conferences (ICLR, ICML, NeurIPS) and associated workshops, covering fluid dynamics, molecular GNNs, and differential equation solvers.
Developed and maintains open-source projects including JAX-SPH (differentiable SPH framework) and LagrangeBench (Lagrangian fluid benchmark suite).
Co-authors
22 total
Co-author 1
Johannes Brandstetter
Johannes Kepler University (JKU) Linz
Co-author 3
Co-author 4
Co-author 5
Johannes Gasteiger, né Klicpera
Anthropic
Co-author 7
Filip Ekström Kelvinius
Linköping University
×
Welcome back
Sign in to Agora
Welcome back! Please sign in to continue.
Email address
Password
Forgot password?
Continue
Do not have an account?
Sign up