Artur Toshev
@ArturToshev
Followers
286
Following
753
Media
13
Statuses
123
PhD student @TU_Muenchen; ML from atoms to fluids
Munich
Joined April 2020
RT @docmilanfar: I AM CALLING FOR A TEMPORARY, BUT TOTAL, SIX-MONTH PAUSE ON ALL NEW PUBLISHING IN AI. I NEED TO CATCH UP WITH THE LITERATU….
0
62
0
RT @jo_brandstetter: Can we model stochastic trajectories of molecules in latent space? No graph, no connectivity, just latent space and sc….
0
5
0
Latent space modeling gone atomistic. Great work @fses91.
Happy to introduce 🔥LaM-SLidE🔥! . We show how trajectories of spatial dynamical systems can be modeled in latent space by. --> leveraging IDENTIFIERS. 📚Paper: .💻Code: 📝Blog: 1/n
0
0
6
RT @O_EA_Carlsson: New paper on Equivariant non-linear maps for neural networks on homogeneous spaces! .We provide steerability constraints….
arxiv.org
This paper presents a novel framework for non-linear equivariant neural network layers on homogeneous spaces. The seminal work of Cohen et al. on equivariant $G$-CNNs on homogeneous spaces...
0
4
0
RT @TsungYiLinCV: DeepSeek R1 demonstrates AI mastering math through reinforcement learning. We introduce Cosmos-Reason1, which learns with….
0
9
0
RT @chaitjo: Introducing All-atom Diffusion Transformers . — towards Foundation Models for generative chemistry, from my internship with th….
0
107
0
@ AI4Science Twitter: consider submitting here 👇.You don't get to present your work next to a Nobel Laureate every day.
📢 Machine Learning Multiscale Processes ICLR 2025 Deadline Extended to Feb 23! 📅. Given low-level theory and computationally-expensive simulation code, how can we model complex systems on a useful time scale? . Why attend?🧵.
0
1
3
RT @kwangmoo_yi: I generally don't like being controversial, but I am sad that this release is completely misleading. It is NOT generative….
0
89
0
RT @oharub: Generating cat videos is nice, but what if you could tackle real scientific problems with the same methods? 🧪🌌.Introducing The….
0
83
0
⚛️ cuEquivariance ⚛️ is something I've been waiting for forever. Thank you @mario1geiger and team!. Curious to see where the "equivariance is just too slow" debate goes 👀.
github.com
cuEquivariance is a math library that is a collective of low-level primitives and tensor ops to accelerate widely-used models, like DiffDock, MACE, Allegro and NEQUIP, based on equivariant neural n...
🔬 Drug and material discovery just got a boost from accelerated computing NVIDIA's new cuEquivariance library accelerates equivariant neural networks by up to 17x. Learn how symmetry & efficiency are reshaping #AI research. #HPC #healthcare .➡️
0
0
15
RT @ask1729: 1/ What are key design principles for scaling neural network interatomic potentials? Our exploration leads us to top results o….
0
36
0
RT @jo_brandstetter: Super hyped to share NeuralDEM -- the first real-time simulation of industrial particulate flows. NeuralDEM replaces D….
0
88
0
RT @gklambauer: Does equivariance matter at scale?. Should a model rather learn equi- and invariances from data or should the architecture….
0
31
0
RT @OpenCatalyst: Introducing Meta’s Open Materials 2024 (OMat24) Dataset and Models! All under permissive open licenses for commercial and….
0
174
0
Latent space modeling beyond the generative AI revolution. Congrats 👏.
Interesting in scaling up neural operators? Happy to announce that Universal Physics Transformers (UPT) -- a scalable framework for neural operators is accepted at #neurips2024. Paper: Project page:
0
0
6
RT @NMcGreivy: Our new paper in @NatMachIntell tells a story about how, and why, ML methods for solving PDEs do not work as well as adverti….
0
110
0
RT @StanfordAILab: arXiv -> alphaXiv. Students at Stanford have built alphaXiv, an open discussion forum for arXiv papers. @askalphaxiv. Yo….
0
2K
0
Happening in about an hour at #201.Happy to see some of you in person!.
🌊 GNNs for particle-based fluid simulations have been struggling to generate stable predictions over long rollouts. Not anymore with Neural SPH!. Accepted at ICML '24. Joint work with @JoErb1337 under the supervision of N. Adams and @jo_brandstetter <3. 1/8
0
1
5
RT @jo_brandstetter: @ArturToshev is an SPH wizard. Particle-based simulations are so interesting - but also challenging. Was really fun to….
0
2
0