mlsanityData to 7 Oct 2026

3D-GSRD: 3D Molecular Graph Auto-Encoder with Selective Re-mask Decoding

arXiv 2510.16780 ↗First posted 19 Oct 2025cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Masked graph modeling (MGM) is a promising approach for molecular representation learning (MRL).However, extending the success of re-mask decoding from 2D to 3D MGM is non-trivial, primarily due to two conflicting challenges: avoiding 2D structure leakage to the decoder, while still providing sufficient 2D context for reconstructing re-masked atoms. To address these challenges, we propose 3D-GSRD: a 3D Molecular Graph Auto-Encoder with Selective Re-mask Decoding. The core innovation of 3D-GSRD lies in its Selective Re-mask Decoding(SRD), which re-masks only 3D-relevant information from encoder representations while preserving the 2D graph structures. This SRD is synergistically integrated with a 3D Relational-Transformer(3D-ReTrans) encoder alongside a structure-independent decoder. We analyze that SRD, combined with the structure-independent decoder, enhances the encoder's role in MRL. Extensive experiments show that 3D-GSRD achieves strong downstream performance, setting a new state-of-the-art on 7 out of 8 targets in the widely used MD17 molecular property prediction benchmark. The code is released at https://github.com/WuChang0124/3D-GSRD.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 55 of its 77 references that are in the map. First posted on arXiv 19 Oct 2025.

Builds on

55 of 77 references.

References · 55

  1. 2017Attention Is All You Needcited by 59,932
  2. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  3. 2018How Powerful are Graph Neural Networks?cited by 4,355
  4. 2020Open Graph Benchmark: Datasets for Machine Learning on Graphscited by 2,148
  5. 2019Strategies for Pre-training Graph Neural Networkscited by 964
  6. 2021E(n) Equivariant Graph Neural Networkscited by 951
  7. 2022Recipe for a General, Powerful, Scalable Graph Transformercited by 702
  8. 2020Benchmarking Graph Neural Networkscited by 672
  9. 2020Directional Message Passing for Molecular Graphscited by 556
  10. 2021Rethinking Graph Transformers with Spectral Attentioncited by 468
  11. 2022GraphMAE: Self-Supervised Masked Graph Autoencoderscited by 443
  12. 2021Equivariant message passing for the prediction of tensorial properties and molecular spectracited by 438
38 more
  1. 2021Graph Neural Networks with Learnable Structural and Positional Representationscited by 326
  2. 2020Learning from Protein Structure with Geometric Vector Perceptronscited by 308
  3. 2020Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Moleculescited by 254
  4. 2021Pre-training Molecular Graph Representation with 3D Geometrycited by 248
  5. 2021Weisfeiler and Lehman Go Cellular: CW Networkscited by 218
  6. 2021ChemRL-GEM: Geometry Enhanced Molecular Representation Learning for Property Predictioncited by 204
  7. 2022TorchMD-NET: Equivariant Transformers for Neural Network based Molecular Potentialscited by 179
  8. 2021GraphiT: Encoding Graph Structure in Transformerscited by 146
  9. 20213D Infomax improves GNNs for Molecular Property Predictioncited by 146
  10. 2020Directional Graph Networkscited by 132
  11. 2023GraphMAE2: A Decoding-Enhanced Masked Self-Supervised Graph Learnercited by 131
  12. 2023MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adaptercited by 128
  13. 2022Pre-training via Denoising for Molecular Property Predictioncited by 124
  14. 2020When Does Self-Supervision Help Graph Convolutional Networks?cited by 121
  15. 2022One Transformer Can Understand Both 2D & 3D Molecular Datacited by 77
  16. 2022ComENet: Towards Complete and Efficient Message Passing for 3D Molecular Graphscited by 69
  17. 2022Molecular Geometry Pretraining with SE(3)-Invariant Denoising Distance Matchingcited by 67
  18. 2021MoCL: Data-driven Molecular Fingerprint via Knowledge-aware Contrastive Learning from Molecular Graphcited by 65
  19. 2024Towards 3D Molecule-Text Interpretation in Language Modelscited by 65
  20. 2021GeomGCL: Geometric Graph Contrastive Learning for Molecular Property Predictioncited by 61
  21. 2021Molecular Contrastive Learning with Chemical Element Knowledge Graphcited by 50
  22. 2023Rethinking Tokenizer and Decoder in Masked Graph Modeling for Moleculescited by 49
  23. 2022Relational Attention: Generalizing Transformers for Graph-Structured Taskscited by 41
  24. 2022Improving Molecular Contrastive Learning via Faulty Negative Mitigation and Decomposed Fragment Contrastcited by 37
  25. 2022Energy-Motivated Equivariant Pretraining for 3D Molecular Graphscited by 37
  26. 2023Fractional Denoising for 3D Molecular Pre-trainingcited by 33
  27. 2024ProtT3: Protein-to-Text Generation for Text-based Protein Understandingcited by 25
  28. 2023Automated 3D Pre-Training for Molecular Property Predictioncited by 24
  29. 2024Uni-Mol2: Exploring Molecular Pretraining Model at Scalecited by 20
  30. 2024UniGEM: A Unified Approach to Generation and Property Prediction for Moleculescited by 20
  31. 2024ReactXT: Understanding Molecular "Reaction-ship" via Reaction-Contextualized Molecule-Text Pretrainingcited by 19
  32. 2020Learn molecular representations from large-scale unlabeled molecules for drug discoverycited by 18
  33. 2023Sliced Denoising: A Physics-Informed Molecular Pre-Training Methodcited by 18
  34. 2025Diffusion Models for Molecules: A Survey of Methods and Taskscited by 18
  35. 2024HIGHT: Hierarchical Graph Tokenization for Molecule-Language Alignmentcited by 17
  36. 2025NExT-Mol: 3D Diffusion Meets 1D Language Modeling for 3D Molecule Generationcited by 17
  37. 2025A Survey of Graph Transformers: Architectures, Theories and Applicationscited by 12
  38. 2025Towards Unified and Lossless Latent Space for 3D Molecular Latent Diffusion Modelingcited by 11

50 most cited of 55 All 55 (JSON)

Built on by

Not cited yet.