# Bridging the Gap between Learning and Inference for Diffusion-Based Molecule Generation

arXiv: https://arxiv.org/abs/2411.05472 · HTML: https://mlsanity.com/paper/2411.05472/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Flow matching, time series and science, one of 14 directions in Diffusion models.
- Builds on 22 of its 48 references that are in the map.
- First posted on arXiv 8 Nov 2024.

## Placed in

- Diffusion models / [Flow matching, time series and science](https://mlsanity.com/diffusion/flow-matching-time-series-and-science/)

## Builds on (most cited first, 22 of 22)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Deep Unsupervised Learning using Nonequilibrium Thermodynamics](https://mlsanity.com/paper/1503.03585/) (2015)
- [GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models](https://mlsanity.com/paper/2112.10741/) (2021)
- [E(n) Equivariant Graph Neural Networks](https://mlsanity.com/paper/2102.09844/) (2021)
- [Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks](https://mlsanity.com/paper/1506.03099/) (2015)
- [Junction Tree Variational Autoencoder for Molecular Graph Generation](https://mlsanity.com/paper/1802.04364/) (2018)
- [SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks](https://mlsanity.com/paper/2006.10503/) (2020)
- [Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation](https://mlsanity.com/paper/1806.02473/) (2018)
- [Professor Forcing: A New Algorithm for Training Recurrent Networks](https://mlsanity.com/paper/1610.09038/) (2016)
- [Pocket2Mol: Efficient Molecular Sampling Based on 3D Protein Pockets](https://mlsanity.com/paper/2205.07249/) (2022)
- [3D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Prediction](https://mlsanity.com/paper/2303.03543/) (2023)
- [A 3D Generative Model for Structure-Based Drug Design](https://mlsanity.com/paper/2203.10446/) (2022)
- [Bridging the Gap between Training and Inference for Neural Machine Translation](https://mlsanity.com/paper/1906.02448/) (2019)
- [Generating 3D Molecules for Target Protein Binding](https://mlsanity.com/paper/2204.09410/) (2022)
- [Elucidating the Exposure Bias in Diffusion Models](https://mlsanity.com/paper/2308.15321/) (2023)
- [Input Perturbation Reduces Exposure Bias in Diffusion Models](https://mlsanity.com/paper/2301.11706/) (2023)
- [DecompDiff: Diffusion Models with Decomposed Priors for Structure-Based Drug Design](https://mlsanity.com/paper/2403.07902/) (2024)
- [Generating 3D Molecules Conditional on Receptor Binding Sites with Deep Generative Models](https://mlsanity.com/paper/2110.15200/) (2021)
- [Pre-training of Graph Neural Network for Modeling Effects of Mutations on Protein-Protein Binding Affinity](https://mlsanity.com/paper/2008.12473/) (2020)
- [Binding-Adaptive Diffusion Models for Structure-Based Drug Design](https://mlsanity.com/paper/2402.18583/) (2024)
- [Recursive Tree Grammar Autoencoders](https://mlsanity.com/paper/2012.02097/) (2020)

## Built on by (most cited first, 1 of 1)

- [Taming the Long Tail: Rebalancing Adversarial Training via Adaptive Perturbation](https://mlsanity.com/paper/2605.13395/) (2026)

## Abstract (arXiv)

The paradigm shift toward structure-driven molecule generation has been propelled by advances in deep generative models, such as variational auto-encoders and diffusion models. However, these generative models for molecular design remain constrained by exposure bias, error accumulation, and suboptimal handling of activity cliffs. Here, we introduce DiffGap, a diffusion-based framework that integrates adaptive sampling and pseudo-molecule estimation to bridge the gap between training objectives and inference dynamics in 3D molecule generation. By dynamically aligning intermediate denoising steps with realistic generation trajectories, DiffGap enables the diffusion model to adapt to input biases in advance during the training phase. A temperature annealing module further controls the aligning strength of the adaptive alignment process, ensuring stable learning of the data distribution. Evaluated on the CrossDocked2020 benchmark, DiffGap outperforms existing methods in docking scores and binding affinity, demonstrating superior fidelity in generating drug-like molecules. Our work establishes a principled approach to harmonize generative training with inference mechanics, offering a robust computational toolkit for accelerating structure-based therapeutic discovery. The source code of DiffGap is available at https://github.com/neusymlab/DiffGap.