Equivariant Denoisers Cannot Copy Graphs: Align Your Graph Diffusion Models
Graph diffusion models, dominant in graph generative modeling, remain underexplored for graph-to-graph translation tasks like chemical reaction prediction. We demonstrate that standard permutation equivariant denoisers face fundamental limitations in these tasks due to their inability to break symmetries in noisy inputs. To address this, we propose aligning input and target graphs to break input symmetries while preserving permutation equivariance in non-matching graph portions. Using retrosynthesis (i.e., the task of predicting precursors for synthesis of a given target molecule) as our application domain, we show how alignment dramatically improves discrete diffusion model performance from 5% to a SOTA-matching 54.7% top-1 accuracy. Code is available at https://github.com/Aalto-QuML/DiffAlign.
Code (1)
Tasks
Chemical Reaction PredictionGraph GenerationGraph-To-Graph TranslationRetrosynthesisMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Equivariant Denoisers for Image Restoration
One key ingredient of image restoration is to define a realistic prior on clean images to complete the missing information in the observation. State-of-the-art restoration methods rely on a neural network to encode this …
DenoisingImage RestorationStochastic OptimizationIsometric Transformation Invariant and Equivariant Graph Convolutional Networks
Graphs are one of the most important data structures for representing pairwise relations between objects. Specifically, a graph embedded in a Euclidean space is essential to solving real problems, such as physical simula…
Object DetectionPhysical SimulationsRoTIR: Rotation-Equivariant Network and Transformers for Fish Scale Image Registration
Image registration is an essential process for aligning features of interest from multiple images. With the recent development of deep learning techniques, image registration approaches have advanced to a new level. In t…
Deep LearningImage RegistrationIt Has Potential: Gradient-Driven Denoisers for Convergent Solutions to Inverse Problems
In recent years there has been increasing interest in leveraging denoisers for solving general inverse problems. Two leading frameworks are regularization-by-denoising (RED) and plug-and-play priors (PnP) which incorpora…
DenoisingE(n) Equivariant Topological Neural Networks
Graph neural networks excel at modeling pairwise interactions, but they cannot flexibly accommodate higher-order interactions and features. Topological deep learning (TDL) has emerged recently as a promising tool for add…
Inductive BiasMolecular Property PredictionProperty Prediction