mlsanityData to 14 Oct 2026

Budgeted Quotient-Residual Guidance for Frozen Pocket-Conditioned Molecular Diffusion

arXiv 2609.31222 ↗First posted 25 Sep 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Pocket-conditioned molecular diffusion updates ambient atom coordinates, but many lead-optimization objectives are expressed on quotient features such as distances, contacts, and anchored substructures. We introduce budgeted quotient-residual guidance (QRG), an inference-time correction that makes these quotient objectives active without retraining the molecular generator. QRG lifts quotient covectors to metric-horizontal ambient directions and delivers them through a trust budget set by the frozen sampler's own step norm: quotient geometry chooses the direction, while sampler motion bounds the scale. We derive the horizontal lift, closed-form sampler-budget update, KL/kinetic interpretation around a frozen reverse step, equivariance conditions, and a product-budget split for budget-capped section and residual controls. Controlled quotient tasks confirm that sampler-relative delivery activates signals that raw local quotient gradients leave dormant. On frozen TargetDiff backbones, official seed-0 CBGBench ligand-generation/editing sweeps show practical quality-runtime gains: Local-QRG improves validity from 0.815 to 0.864 on fragment growing, 0.664 to 0.707 on scaffold hopping, and 0.681 to 0.712 on linker design, while PredNext-QRG improves fragment/scaffold and remains near-neutral on linker. Novelty remains 1.000 and diversity is preserved in the matched multi-seed molecular slice, giving task-dependent improvements without sampler retraining or backbone modification. Overall, QRG provides a lightweight route to quotient-aware inference for frozen molecular samplers with explicit runtime accounting.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 45 of its 56 references that are in the map. First posted on arXiv 25 Sep 2026.

Builds on

45 of 56 references.

References · 45

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  3. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  4. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  5. 2022Classifier-Free Diffusion Guidancecited by 5,847
  6. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,101
  7. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  8. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 3,045
  9. 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,430
  10. 2022Diffusion Posterior Sampling for General Noisy Inverse Problemscited by 1,416
  11. 2021Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gaugescited by 1,167
  12. 2016Group Equivariant Convolutional Networkscited by 1,165
33 more
  1. 2021Variational Diffusion Modelscited by 1,153
  2. 2021E(n) Equivariant Graph Neural Networkscited by 965
  3. 2022Denoising Diffusion Restoration Modelscited by 963
  4. 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 746
  5. 2021E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentialscited by 708
  6. 2022Equivariant Diffusion for Molecule Generation in 3Dcited by 677
  7. 2020SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networkscited by 607
  8. 2021Diffusion Schr"odinger Bridge with Applications to Score-Based Generative Modelingcited by 602
  9. 2022GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generationcited by 516
  10. 2023Universal Guidance for Diffusion Modelscited by 437
  11. 20183D Steerable CNNs: Learning Rotationally Equivariant Features in Volumetric Datacited by 354
  12. 2022DiffDock: Diffusion Steps, Twists, and Turns for Molecular Dockingcited by 313
  13. 2018On the Generalization of Equivariance and Convolution in Neural Networks to the Action of Compact Groupscited by 300
  14. 2018GuacaMol: Benchmarking Models for De Novo Molecular Designcited by 297
  15. 2013A survey of the Schr"odinger problem and some of its connections with optimal transportcited by 292
  16. 2018Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Modelscited by 275
  17. 2019Cormorant: Covariant Molecular Neural Networkscited by 263
  18. 2022Torsional Diffusion for Molecular Conformer Generationcited by 263
  19. 2020Equivariant Flows: Exact Likelihood Generative Learning for Symmetric Densitiescited by 250
  20. 2020Generalizing Convolutional Neural Networks for Equivariance to Lie Groups on Arbitrary Continuous Datacited by 248
  21. 2023SE(3) diffusion model with application to protein backbone generationcited by 219
  22. 2022Structure-based Drug Design with Equivariant Diffusion Modelscited by 217
  23. 2022Pocket2Mol: Efficient Molecular Sampling Based on 3D Protein Pocketscited by 183
  24. 20233D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Predictioncited by 177
  25. 2019Symmetry-adapted generation of 3d point sets for the targeted discovery of moleculescited by 172
  26. 2023SE(3)-Stochastic Flow Matching for Protein Backbone Generationcited by 150
  27. 20213D Infomax improves GNNs for Molecular Property Predictioncited by 146
  28. 2023Equivariant flow matchingcited by 135
  29. 2022Equivariant 3D-Conditional Diffusion Models for Molecular Linker Designcited by 122
  30. 2024DecompDiff: Diffusion Models with Decomposed Priors for Structure-Based Drug Designcited by 88
  31. 2020Generating 3D Molecular Structures Conditional on a Receptor Binding Site with Deep Generative Modelscited by 33
  32. 2024CBGBench: Fill in the Blank of Protein-Molecule Complex Binding Graphcited by 19
  33. 2025Rao-Blackwell Gradient Estimators for Equivariant Denoising Diffusioncited by 5

Built on by

Not cited yet.