Related papers: Molecular Representations in Implicit Functional Space via Hyper-Networks

Molecular Representations in Implicit Functional Space via Hyper-Networks

URL: http://arxiv.org/abs/2601.22327v1
Date: Thu, 29 Jan 2026 21:13:37 GMT
Title: Molecular Representations in Implicit Functional Space via Hyper-Networks
Authors: Zehong Wang, Xiaolong Han, Qi Yang, Xiangru Tang, Fang Wu, Xiaoguang Guo, Weixiang Sun, Tianyi Ma, Pietro Lio, Le Cong, Sheng Wang, Chuxu Zhang, Yanfang Ye,
Abstract summary: We argue that molecular learning can instead be formulated as learning in function space.<n>We instantiate this formulation with MolField, a hyper-network-based framework that learns distributions over molecular fields.<n>Our results show that treating molecules as continuous functions fundamentally changes how molecular representations generalize across tasks.
Score: 53.70982267248536
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: Molecular representations fundamentally shape how machine learning systems reason about molecular structure and physical properties. Most existing approaches adopt a discrete pipeline: molecules are encoded as sequences, graphs, or point clouds, mapped to fixed-dimensional embeddings, and then used for task-specific prediction. This paradigm treats molecules as discrete objects, despite their intrinsically continuous and field-like physical nature. We argue that molecular learning can instead be formulated as learning in function space. Specifically, we model each molecule as a continuous function over three-dimensional (3D) space and treat this molecular field as the primary object of representation. From this perspective, conventional molecular representations arise as particular sampling schemes of an underlying continuous object. We instantiate this formulation with MolField, a hyper-network-based framework that learns distributions over molecular fields. To ensure physical consistency, these functions are defined over canonicalized coordinates, yielding invariance to global SE(3) transformations. To enable learning directly over functions, we introduce a structured weight tokenization and train a sequence-based hyper-network to model a shared prior over molecular fields. We evaluate MolField on molecular dynamics and property prediction. Our results show that treating molecules as continuous functions fundamentally changes how molecular representations generalize across tasks and yields downstream behavior that is stable to how molecules are discretized or queried.

Related papers

HyperDiffusionFields (HyDiF): Diffusion-Guided Hypernetworks for Learning Implicit Molecular Neural Fields [12.849722578846178]
We introduce HyperDiffusionFields (HyDiF), a framework that models 3D molecular conformers as continuous fields.<n>At the core of our approach is the Molecular Directional Field (MDF), a vector field that maps any point in space to the direction of the nearest atom of a particular type.<n>We demonstrate that our approach scales to larger biomolecules, illustrating a promising direction for field-based molecular modeling.
arXiv Detail & Related papers (2025-10-20T21:41:10Z)
Swallowing the Bitter Pill: Simplified Scalable Conformer Generation [12.341835649897886]
We present a novel way to predict molecular conformers through a simple formulation that sidesteps many of the equis of prior works and achieves state of the art results by using the advantages of scale. We are able to radically simplify structure learning, and make it trivial to scale up the model sizes. This model, called Molecular Conformer Fields (MCF), works by parameterizing conformer structures as functions that map elements from a molecular graph directly to their 3D location in space.
arXiv Detail & Related papers (2023-11-27T22:53:41Z)
Molecule Design by Latent Space Energy-Based Modeling and Gradual Distribution Shifting [53.44684898432997]
Generation of molecules with desired chemical and biological properties is critical for drug discovery. We propose a probabilistic generative model to capture the joint distribution of molecules and their properties. Our method achieves very strong performances on various molecule design tasks.
arXiv Detail & Related papers (2023-06-09T03:04:21Z)
Towards Predicting Equilibrium Distributions for Molecular Systems with Deep Learning [60.02391969049972]
We introduce a novel deep learning framework, called Distributional Graphormer (DiG), in an attempt to predict the equilibrium distribution of molecular systems. DiG employs deep neural networks to transform a simple distribution towards the equilibrium distribution, conditioned on a descriptor of a molecular system.
arXiv Detail & Related papers (2023-06-08T17:12:08Z)
MUDiff: Unified Diffusion for Complete Molecule Generation [104.7021929437504]
We present a new model for generating a comprehensive representation of molecules, including atom features, 2D discrete molecule structures, and 3D continuous molecule coordinates. We propose a novel graph transformer architecture to denoise the diffusion process. Our model is a promising approach for designing stable and diverse molecules and can be applied to a wide range of tasks in molecular modeling.
arXiv Detail & Related papers (2023-04-28T04:25:57Z)
Modeling Molecular Structures with Intrinsic Diffusion Models [2.487445341407889]
This thesis proposes Intrinsic Diffusion Modeling. It combines diffusion generative models with scientific knowledge about the flexibility of biological complexes. We demonstrate the effectiveness of this approach on two fundamental tasks at the basis of computational chemistry and biology.
arXiv Detail & Related papers (2023-02-23T03:26:48Z)
DiffBP: Generative Diffusion of 3D Molecules for Target Protein Binding [51.970607704953096]
Previous works usually generate atoms in an auto-regressive way, where element types and 3D coordinates of atoms are generated one by one. In real-world molecular systems, the interactions among atoms in an entire molecule are global, leading to the energy function pair-coupled among atoms. In this work, a generative diffusion model for molecular 3D structures based on target proteins is established, at a full-atom level in a non-autoregressive way.
arXiv Detail & Related papers (2022-11-21T07:02:15Z)
Molecular dynamics without molecules: searching the conformational space of proteins with generative neural networks [0.0]
All-atom and coarse-grained molecular dynamics are widely used to study the conformational states of proteins. All-atom and coarse-grained simulation methods suffer from the fact that without access to supercomputing resources, the time and length scales at which these states become detectable are difficult to achieve. One alternative is based on encoding the atomistic trajectory of molecular dynamics as a shorthand version of physical particles, and then learning to propagate the encoded trajectory through the use of artificial intelligence vectors.
arXiv Detail & Related papers (2022-06-09T02:06:43Z)
Scalable Fragment-Based 3D Molecular Design with Reinforcement Learning [68.8204255655161]
We introduce a novel framework for scalable 3D design that uses a hierarchical agent to build molecules. In a variety of experiments, we show that our agent, guided only by energy considerations, can efficiently learn to produce molecules with over 100 atoms.
arXiv Detail & Related papers (2022-02-01T18:54:24Z)
Learning a Continuous Representation of 3D Molecular Structures with Deep Generative Models [0.0]
Generative models are an entirely different approach that learn to represent and optimize molecules in a continuous latent space. We describe deep generative models of three dimensional molecular structures using atomic density grids. We are also able to sample diverse sets of molecules based on a given input compound to increase the probability of creating valid, drug-like molecules.
arXiv Detail & Related papers (2020-10-17T01:15:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.