[Upstream sync] ai-boost/awesome-ai-for-science (github) — 0 added, 1 modified #84

Open
promptadmin wants to merge 1 commits from upstream-sync/awesome-ai-for-science-20260822-ca4bd5-roew into main
@@ -2,9 +2,9 @@
title: "Readme"
task: ""
lineage_type: import
upstream_source: https://github.com/ai-boost/awesome-ai-for-science/blob/ec6f3650/README.md
upstream_sha: ec6f3650
imported_at: 2026-08-21
upstream_source: https://github.com/ai-boost/awesome-ai-for-science/blob/ca4bd58e/README.md
upstream_sha: ca4bd58e
imported_at: 2026-08-22
prompt_class: catalogue
upstream_changes: accepted
author: upstream
@@ -744,6 +744,7 @@ validated: false
- [Neural ODEs](https://github.com/rtqichen/torchdiffeq) - Differential equations with neural networks
- [Physics-Informed Neural Networks](https://github.com/maziarraissi/PINNs) - Physics-constrained ML
- [EquiformerV2](https://github.com/atomicarchitects/equiformer_v2) - Improved equivariant Transformer for 3D atomic graphs (ICLR2024)
- [EquiformerV3](https://github.com/atomicarchitects/equiformer_v3) - Scaling efficient, expressive, and general SE(3)-equivariant graph attention transformers for atomic systems and machine-learned interatomic potentials (MIT License, 2026)
- [Equiformer](https://github.com/atomicarchitects/equiformer) - Equivariant graph attention Transformer (ICLR2023)
- [TORAX](https://github.com/google-deepmind/torax) - Differentiable tokamak core transport simulator for fusion energy research, coupling PDE solvers with JAX auto-differentiation and neural-network surrogates for fast forward modelling, pulse-design, and trajectory optimization (Google DeepMind, Apache 2.0)
- [DiffPhysDrone (Nature Machine Intelligence 2025)](https://github.com/HenryHuYu/DiffPhysDrone) - First real quadrotor robot trained end-to-end with differentiable physics for vision-based agile flight, bridging simulation-based learning and real-world deployment with physics-informed neural network controllers (558+ stars)