Research map / Image generation / Autoregressive and latent generation
Invariant Conditional Molecular Generation Under Distribution Shift
Chunyu Hu, Tianyin Liao, Yicheng Sui, Ran Zhang, Xiao Wang, Ziwei Zhang
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 structural consistency loss Lcons (ablation) | IC-MOL · QM9 HOMO (S->L shift) | MAE: 613.6 -> 611.4 | Improved |
| add structural consistency loss Lcons (ablation) | IC-MOL · QM9 HOMO (S->L shift) 3D | molecule stability: 0.941 -> 0.985 | Improved |
| add structural consistency loss Lcons (ablation) | IC-MOL · QM9 HOMO (S->L shift) 2D | Frag: 0.812 -> 0.841 | Improved |
| add cross-attention (ablation) | IC-MOL · QM9 HOMO (S->L shift) | MAE: 615.8 -> 611.4 | Improved |
| add cross-attention (ablation) | IC-MOL · QM9 HOMO (S->L shift) 3D | molecule stability: 0.966 -> 0.985 | Improved |
| add cross-attention (ablation) | IC-MOL · QM9 HOMO (S->L shift) 2D | validity: 0.965 -> 0.991 | Improved |
| add second phase (ablation) | IC-MOL · QM9 HOMO (S->L shift) 3D | molecule stability: 0.826 -> 0.985 | Improved |
| add second phase (ablation) | IC-MOL · QM9 HOMO (S->L shift) | MAE: 612.4 -> 611.4 | Improved |
| add first phase (invariant molecule generation) (ablation) | IC-MOL · QM9 HOMO (S->L shift) 3D | molecule stability: 0.911 -> 0.985 | Improved |
| add first phase (invariant molecule generation) (ablation) | IC-MOL · QM9 HOMO (S->L shift) | MAE: 612.6 -> 611.4 | Improved |
| IC-MOL instead of JODO under S->L shift | JODO · QM9 Cv (S->L shift) | MAE: 1.235 -> 1.209 | Improved |
| add DRO to JODO | JODO · QM9 Cv (S->L shift) | MAE: 1.235 -> 1.594 | Hurt |
Closest papers by meaning
- Lift Your Molecules: Molecular Graph Generation in Latent Euclidean Space (ICLR 2025 · similarity 0.89)
- Geometric Representation Condition Improves Equivariant Molecule Generation (ICML 2025 · similarity 0.88)
- Graph Diffusion Transformers for Multi-Conditional Molecular Generation (NeurIPS 2024 · similarity 0.88)
- Learning Disentangled Equivariant Representation for Explicitly Controllable 3D Molecule Generation (AAAI 2025 · similarity 0.87)
- Controlled Generation with Equivariant Variational Flow Matching (ICML 2025 · similarity 0.87)
- DMol: A Highly Efficient and Chemical Motif-Preserving Molecule Generation Platform (NeurIPS 2025 · similarity 0.87)
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.
