# Multi-level Self-supervised Pretraining on Compositional Hierarchical Graph for Molecular Property Prediction

arXiv: https://arxiv.org/abs/2605.16088 · HTML: https://mlsanity.com/paper/2605.16088/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 41 references that are in the map.
- First posted on arXiv 15 May 2026.

## Builds on (most cited first, 19 of 19)

- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [How Powerful are Graph Neural Networks?](https://mlsanity.com/paper/1810.00826/) (2018)
- [Graph Contrastive Learning with Augmentations](https://mlsanity.com/paper/2010.13902/) (2020)
- [Deep Graph Infomax](https://mlsanity.com/paper/1809.10341/) (2018)
- [MoleculeNet: A Benchmark for Molecular Machine Learning](https://mlsanity.com/paper/1703.00564/) (2017)
- [Strategies for Pre-training Graph Neural Networks](https://mlsanity.com/paper/1905.12265/) (2019)
- [Analyzing Learned Molecular Representations for Property Prediction](https://mlsanity.com/paper/1904.01561/) (2019)
- [Graph Contrastive Learning Automated](https://mlsanity.com/paper/2106.07594/) (2021)
- [Motif-based Graph Self-Supervised Learning for Molecular Property Prediction](https://mlsanity.com/paper/2110.00987/) (2021)
- [Molecule Generation by Principal Subgraph Mining and Assembling](https://mlsanity.com/paper/2106.15098/) (2021)
- [Rethinking Tokenizer and Decoder in Masked Graph Modeling for Molecules](https://mlsanity.com/paper/2310.14753/) (2023)
- [Fragment-based Pretraining and Finetuning on Molecular Graphs](https://mlsanity.com/paper/2310.03274/) (2023)
- [Motif-aware Attribute Masking for Molecular Graph Pre-training](https://mlsanity.com/paper/2309.04589/) (2023)
- [Pre-training Graph Neural Networks on Molecules by Using Subgraph-Conditioned Graph Information Bottleneck](https://mlsanity.com/paper/2412.15589/) (2024)
- [SGAC: A Graph Neural Network Framework for Imbalanced and Structure-Aware AMP Classification](https://mlsanity.com/paper/2412.16276/) (2024)
- [Improving Self-supervised Molecular Representation Learning using Persistent Homology](https://mlsanity.com/paper/2311.17327/) (2023)
- [FragNet: A Graph Neural Network for Molecular Property Prediction with Four Levels of Interpretability](https://mlsanity.com/paper/2410.12156/) (2024)
- [Enhancing Molecular Property Predictions by Learning from Bond Modelling and Interactions](https://mlsanity.com/paper/2603.00568/) (2026)
- [Hierarchical Molecular Representation Learning via Fragment-Based Self-Supervised Embedding Prediction](https://mlsanity.com/paper/2602.20344/) (2026)

## Abstract (arXiv)

Self-supervised pretraining on molecular graphs has emerged as a promising approach for molecular property prediction, yet most existing methods operate at a single structural granularity and treat bond information as auxiliary edge attributes rather than as an independent semantic layer. In this work, we propose MolCHG, a multi-level self-supervised pretraining framework built upon a novel Compositional Hierarchical Graph that organizes molecular structure into four types of nodes across three semantic levels. By introducing a bond graph that operates in parallel with the atom graph, our architecture elevates bond-level information to independently evolving node representations, enabling fragment nodes to aggregate atom-level and bond-level semantics on an equal footing. We design three level-specific pretraining objectives: an atom-bond cross-view contrastive task that aligns the atom-view and bond-view representations within each fragment, a fragment-level functional group prediction task to inject domain-relevant chemical knowledge, and graph-level structure prediction tasks to encode global molecular topology. Experiments on nine MoleculeNet benchmarks demonstrate that MolCHG achieves the best performance on seven datasets across both classification and regression tasks, remaining competitive with the strongest baselines on the rest. Ablation studies further confirm that the multi-level supervision signals are complementary and that each component contributes to the overall performance.