Research map / Probabilistic and scientific modeling / Molecules and proteins
Learning Molecular Chirality via Chiral Determinant Kernels
Runhan Shi, Zhicheng Zhang, Letian Chen, Gufeng Yu, Yang Yang
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 |
|---|---|---|---|
| feed the chirality matrix into the chiral encoder (ablation) | ChiDeK · ACMP axial ECD peak symbol | accuracy (%): 51.2 -> 68.7 | Improved |
| use determinant-kernel encoder instead of linear encoder with chirality matrix (ablation) | ChiDeK · ACMP axial ECD peak symbol | accuracy (%): 68.7 -> 71.2 | Improved |
| add QR on weights to guarantee full column rank (ablation) | ChiDeK · ACMP axial ECD peak symbol | accuracy (%): 68.3 -> 71.2 | Improved |
| use weight regularization instead of QR for rank (ablation) | ChiDeK · ACMP axial ECD peak symbol | accuracy (%): 71.2 vs 71.0 | No clear effect |
| separate chiral-related atoms from non-chiral atoms in attention keys/values (ablation) | ChiDeK · ACMP axial ECD peak symbol | accuracy (%): 69.8 -> 71.2 | Improved |
| ChiDeK vs best baseline on axial OR prediction | ChiDeK · ACMP axial OR | rotation accuracy (%): 65.0 -> 69.2 | Improved |
| ChiDeK vs best baseline on axial ECD symbol prediction | ChiDeK · ACMP axial ECD | peak symbol accuracy (%): 60.4 -> 71.2 | Improved |
| ChiDeK vs ChIRo on enantiomer ranking | ChiDeK · Enantiomer ranking | accuracy (%): 72.0 -> 72.8 | Improved |
| ChiDeK vs ECDFormer on central ECD peak position/number | ChiDeK · CMCDS ECD | RMSE (position): 2.02 -> 2.20 | Hurt |
Closest papers by meaning
- Learning 3D Representations of Molecular Chirality with Invariance to Bond Rotations (ICLR 2022 · similarity 0.93)
- E(3)-equivariant models cannot learn chirality: Field-based molecular generation (ICLR 2025 · similarity 0.90)
- Learning Over Molecular Conformer Ensembles: Datasets and Benchmarks (ICLR 2024 · similarity 0.86)
- A Plug-and-Play Quaternion Message-Passing Module for Molecular Conformation Representation (AAAI 2024 · similarity 0.84)
- TRIDENT: Tri-Modal Molecular Representation Learning with Taxonomic Annotations and Local Correspondence (NeurIPS 2025 · similarity 0.84)
- HAGO-Net: Hierarchical Geometric Message Passing for Molecular Representation Learning (AAAI 2024 · 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.
