The Year in AI Papers
The year/Independent research

Paper 2602.17602

MolHIT: Advancing Molecular-Graph Generation with Hierarchical Discrete Diffusion Models

Hojung Jung, Rodrigo Hormazabal, Jaehyeong Jo, Youngrok Park, Kyunggeun Roh, Se-Young Yun, Sehui Han, Dae-Woong Jeong

Published
Feb 2026
Research lab
Independent
Citations
0
GitHub
Not linked

01 In brief

Summary

MolHIT is a molecular graph generation framework based on a Hierarchical Discrete Diffusion Model (HDDM) that generalizes discrete diffusion by adding mid-level states encoding chemical priors, and a Decoupled Atom Encoding (DAE) that splits atom types by chemical roles (aromaticity, formal charge).

This addresses limitations in prior graph diffusion models, which suffer from low validity and ill-posed atom encoding.

MolHIT achieves state-of-the-art performance on the MOSES dataset with near-perfect validity (99.1%) and quality (94.2%), surpassing both 2D graph and 1D sequence baselines.

It also shows strong results on GuacaMol, multi-property guided generation, and scaffold extension.

Key components include a Project-and-Noise (PN) sampler and temperature sampling for quality-diversity trade-offs.

The framework is validated with a mathematically guaranteed ELBO and ablation studies showing each component contributes to performance gains.

02 From the paper

Abstract

Molecular generation with diffusion models has emerged as a promising direction for AI-driven drug discovery and materials science. While graph diffusion models have been widely adopted due to the discrete nature of 2D molecular graphs, existing models suffer from low chemical validity and struggle to meet the desired properties compared to 1D modeling. In this work, we introduce MolHIT, a powerful molecular graph generation framework that overcomes long-standing performance limitations in existing methods. MolHIT is based on the Hierarchical Discrete Diffusion Model, which generalizes discrete diffusion to additional categories that encode chemical priors, and decoupled atom encoding that splits the atom types according to their chemical roles. Overall, MolHIT achieves new state-of-the-art performance on the MOSES dataset with near-perfect validity for the first time in graph diffusion, surpassing strong 1D baselines across multiple metrics. We further demonstrate strong performance in downstream tasks, including multi-property guided generation and scaffold extension.