Research map / Probabilistic and scientific modeling / Differential equations and PDEs
MicroEvoEval: A Systematic Evaluation Framework for Image-Based Microstructure Evolution Prediction
Qinyi Zhang, Duanyu Feng, Ronghui Han, Yangshuai Wang, Hao Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use VMamba architecture instead of PredRNN for long-term rollout | microstructure prediction · Grain growth (10-90) | SSIM: 0.26338 -> 0.97322 | Improved |
| use VMamba instead of PredRNN | microstructure prediction · Dendritic solidification (10-90) | SSIM: 0.43134 -> 0.97682 | Improved |
| use VMamba architecture | microstructure prediction · Grain growth | inference time (s): 0.082 -> 0.006 | Improved |
| rank models by short-term accuracy to predict long-term stability | benchmark · Plane-wave propagation | RMSE: VRMNN best short RMSE 0.00081; long-term advantage diminishes | Hurt |
| select model by numerical accuracy (RMSE/SSIM) instead of physical fidelity | benchmark · Spinodal decomposition (10-90) | L-ETAP: VMamba -1.944 vs TAU -1.958 | Mixed |
| use VMamba (Mamba) vs older RNNs such as ConvLSTM | microstructure prediction · Plane-wave propagation (10-90) | SSIM: 0.54013 -> 0.98745 | Improved |
| use VMRNN (Mamba-LSTM hybrid) vs PredFormer | general spatio-temporal models · Grain growth (short-term) | RMSE: 0.02453 -> 0.02230 | Improved |
| domain-specific vs general-purpose architecture | benchmark · Long-term 10-90 | SSIM | No clear effect |
Closest papers by meaning
- Co-PiLOT: Constrained Physics-Informed Latent Optimization for Target-Driven Inverse Design (NeurIPS 2026 · similarity 0.87)
- Scalable Diffusion for Materials Generation (ICLR 2024 · similarity 0.86)
- LEIA: Learned Environment for Interactive Architected Materials (NeurIPS 2026 · similarity 0.85)
- Vector Field Oriented Diffusion Model for Crystal Material Generation (AAAI 2024 · similarity 0.85)
- Offline Materials Optimization with CliqueFlowmer (NeurIPS 2026 · similarity 0.85)
- Diffusion-Based Hierarchical Graph Neural Networks for Simulating Nonlinear Solid Mechanics (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
