Research map / Probabilistic and scientific modeling / Molecules and proteins
Can Molecular Evolution Mechanism Enhance Molecular Representation?
Kun Li, Longtao Hu, Jiameng Chen, Hongzhi Zhang, Yida Xiong, Xiantao Cai, Wenbin Hu, Jia Wu
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 |
|---|---|---|---|
| add MEvoN-MPP to GCN | GCN · QM7 | MAE: 122.9 -> 45.9 | Improved |
| add MEvoN-MPP to Equiformer | Equiformer · QM7 | MAE: 58.3 -> 55.5 | Improved |
| add MEvoN-MPP to ViSNet | ViSNet · QM7 | MAE: 69.8 -> 56.9 | Improved |
| add MEvoN-MPP to Equiformer | Equiformer · QM9 HOMO | MAE: 0.0263 -> 0.1156 (Evo-branch), 0.0223 (Mol-branch) | Mixed |
| add MEvoN-MPP to GCN | GCN · QM9 gap | MAE: 0.5928 -> 0.2579 | Improved |
| add MEvoN-MPP across backbones on QM9 | GCN, GIN, SchNet, ComENet, Equiformer, ViSNet · QM9 (8 properties) | MAE: average 32.3% improvement | Improved |
| add EvoP module only (no EvoL) (ablation) | MEvoN-MPP · QM9 gap | MAE: 0.2916 -> 0.9809 | Hurt |
| add EvoL module only (no EvoP) (ablation) | MEvoN-MPP · QM9 gap | MAE: 0.2916 -> 0.4051 | Hurt |
| combine EvoP and EvoL (ablation) | MEvoN-MPP · QM9 gap | MAE: 0.2916 -> 0.2270 | Improved |
| vary number of evolutionary paths K (1 to 9) (ablation) | MEvoN-MPP · QM9 HOMO | regression metrics | Mixed |
Closest papers by meaning
- Representing Molecules as Random Walks Over Interpretable Grammars (ICML 2024 · similarity 0.85)
- Efficient Evolutionary Search Over Chemical Space with Large Language Models (ICLR 2025 · similarity 0.83)
- Atomas: Hierarchical Adaptive Alignment on Molecule-Text for Unified Molecule Understanding and Generation (ICLR 2025 · similarity 0.83)
- Searching for High-Value Molecules Using Reinforcement Learning and Transformers (ICLR 2024 · similarity 0.83)
- Learning Over Molecular Conformer Ensembles: Datasets and Benchmarks (ICLR 2024 · similarity 0.82)
- From Artificially Real to Real: Leveraging Pseudo Data from Large Language Models for Low-Resource Molecule Discovery (AAAI 2024 · similarity 0.82)
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.
