Research map / Probabilistic and scientific modeling / Molecules and proteins
CARD: Coarse-to-fine Autoregressive Modeling with Radix-based Decomposition for Transferable Free Energy Estimation
Ziyang Yu, Yi He, Wenbing Huang, Wen Yan, Yang Liu
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 Stage II training with force-field energy alignment loss (ablation) | CARD · Vacuum-to-toluene solvation | MAE (kcal/mol): 0.81 -> 0.71 | Improved |
| use radix 4 instead of 3 (ablation) | CARD (L=3, Stage I) · Vacuum-to-toluene solvation | MAE (kcal/mol): 2.43 -> 0.81 | Improved |
| use radix 4 instead of 5 (ablation) | CARD (L=3, Stage I) · Vacuum-to-toluene solvation | MAE (kcal/mol): 1.88 -> 0.81 | Improved |
| increase depth from L=2 to L=3 (coarse-to-fine levels) (ablation) | CARD (r=4, Stage I) · Vacuum-to-toluene solvation | MAE (kcal/mol): 5.85 -> 0.81 | Improved |
| increase depth from L=3 to L=4 (ablation) | CARD (r=4, Stage I) · Vacuum-to-toluene solvation | MAE (kcal/mol): 0.81 -> 1.43 | Hurt |
| use CARD zero-free-energy proposal instead of DFT energy difference between minima | aqueous tautomer free energy · 27 tautomer pairs | MAE (kcal/mol): 4.62 -> 4.11 | Improved |
| use CARD instead of regression-based sPhysNet-pre | aqueous tautomer free energy · 27 tautomer pairs | Pearson correlation: 0.35 -> 0.64 | Improved |
| use CARD instead of MFES for solvation free energy | inference time · 70 solvation test systems | inference time (s): 3.23 × 104 -> 7.70 × 102 (excluding endstate MD) | Improved |
Closest papers by meaning
- Aligning Transformers with Continuous Feedback via Energy Rank Alignment (NeurIPS 2025 · similarity 0.86)
- Action-Minimization Meets Generative Modeling: Efficient Transition Path Sampling with the Onsager-Machlup Functional (ICML 2025 · similarity 0.86)
- Generative Modeling of Molecular Dynamics Trajectories (NeurIPS 2024 · similarity 0.86)
- UniSim: A Unified Simulator for Time-Coarsened Dynamics of Biomolecules (ICML 2025 · similarity 0.85)
- Navigating the Design Space of Equivariant Diffusion-Based Generative Models for De Novo 3D Molecule Generation (ICLR 2024 · similarity 0.85)
- FlashMD: long-stride, universal prediction of molecular dynamics (NeurIPS 2025 · similarity 0.85)
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.
