---
title: "FACT — paper by Nam et al. (2026) · BioNLP 2026"
description: "SMILES-based molecular encoders usually borrow random masked language modeling from NLP, which ignores chemical… 29 references."
authors: ["Hyeonyeong Nam", "Woojae Choi", "Deok-Joong Lee", "Young-Han Son", "Sangwoon Lee", "Bogyeong Kang", "Eunjung Jo", "Tae-Eui Kam"]
year: 2026
venue: "BioNLP 2026"
source: "https://aclanthology.org/2026.bionlp-1.56/"
doi: "10.18653/v1/2026.bionlp-1.56"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/nam-2026-fact-functional-group-alignment-consistency/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/nam-2026-fact-functional-group-alignment-consistency/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/33/) / FACT: Functional Group Alignment and Consistency in Token Space for Structure-aware Molecular Representation Learning

# FACT: Functional Group Alignment and Consistency in Token Space for Structure-aware Molecular Representation Learning

**Authors:** Hyeonyeong Nam, Woojae Choi, Deok-Joong Lee, Young-Han Son, Sangwoon Lee, Bogyeong Kang, Eunjung Jo and Tae-Eui Kam · **Year:** 2026 · **Venue:** [BioNLP 2026](https://md.research.elamur.ai/free/en/venues/bionlp/) ([BioNLP 2026](https://md.research.elamur.ai/free/en/venues/bionlp/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.bionlp-1.56/) · [PDF](https://aclanthology.org/2026.bionlp-1.56.pdf) · [DOI 10.18653/v1/2026.bionlp-1.56](https://doi.org/10.18653/v1/2026.bionlp-1.56) · [OpenAlex](https://openalex.org/W7166796708)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
SMILES-based molecular encoders usually borrow random masked language modeling from NLP, which ignores chemical structure. Earlier functional-group-aware methods (FG-BERT, MLM-FG) try to fix this, but they leave two gaps. Functional groups are defined over graph atoms, while SMILES mixes atom tokens with bond, branch and ring-closure symbols, so masking hides only part of each group. Also, one molecule has many valid SMILES strings, so the same group can sit at different token positions.

FACT addresses both gaps. It finds functional groups without predefined patterns, using the Ertl algorithm through the EFGs framework. It then uses RDKit's atom output order to map graph atoms to SMILES tokens deterministically, and widens each group to a contiguous span that includes adjacent structural tokens. Pre-training combines full FG span masking with random masking. Fine-tuning adds a consistency loss that keeps FG representations aligned across two SMILES forms of the same molecule.

On eight MoleculeNet tasks, FACT reaches state-of-the-art or competitive results. This suggests that structure-aware tokenization alignment can improve sequence-based molecular models.

## Key points
- Functional groups are detected with the pattern-free Ertl algorithm (via EFGs) instead of SMARTS templates, and atom–token alignment relies on RDKit's atom output order.
- Hybrid pre-training masks complete FG token spans, covering both atoms and structural tokens, together with random tokens. The fine-tuning stage concatenates global and FG-level embeddings from two SMILES views and adds an FG-preserving consistency loss.
- On regression tasks, FACT is best on FreeSolv but falls behind the top baselines on ESOL and Lipophilicity. The authors link this to these properties depending more on global molecular features.
- A UMAP analysis of 10,000 PubChem molecules shows tighter grouping of molecules sharing a functional group than MLM-FG produces. Layer Integrated Gradients on BACE (62 FACT-only-correct, 53 both-correct, 18 both-wrong cases) show attributions concentrated on true functional groups.
- The model is released on Hugging Face as `Zaeus/FACT`.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Korea University](https://md.research.elamur.ai/free/en/universities/korea-university/) — University, Seoul, South Korea · 29 works · 9,490 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🧮 Computational Theory and Mathematics](https://md.research.elamur.ai/free/en/shelves/computer-science/computational-theory-and-mathematics/) · 321 works · [🧱 Materials Science](https://md.research.elamur.ai/free/en/shelves/materials-science/) · 31 works · [🧱 Materials Chemistry](https://md.research.elamur.ai/free/en/shelves/materials-science/materials-chemistry/) · 31 works · [🧪 Chemistry](https://md.research.elamur.ai/free/en/shelves/chemistry/) · 4 works · [🧪 Physical and Theoretical Chemistry](https://md.research.elamur.ai/free/en/shelves/chemistry/physical-and-theoretical-chemistry/) · 4 works

## Cite

Cite as: Hyeonyeong Nam, Woojae Choi, Deok-Joong Lee, Young-Han Son, Sangwoon Lee, Bogyeong Kang, Eunjung Jo and Tae-Eui Kam. 2026. FACT: Functional Group Alignment and Consistency in Token Space for Structure-aware Molecular Representation Learning. BioNLP 2026.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{nam2026fact,
  title = {{FACT: Functional Group Alignment and Consistency in Token Space for Structure-aware Molecular Representation Learning}},
  author = {Hyeonyeong Nam, Woojae Choi, Deok-Joong Lee, Young-Han Son, Sangwoon Lee, Bogyeong Kang, Eunjung Jo and Tae-Eui Kam},
  booktitle = {BioNLP 2026},
  year = {2026},
  doi = {10.18653/v1/2026.bionlp-1.56},
  url = {https://doi.org/10.18653/v1/2026.bionlp-1.56}
}
```

- [← previous Facilitating Large Language Model-based Multiple-choice Reading Comprehension with Question-solving Thought](https://md.research.elamur.ai/free/en/papers/su-2026-facilitating-large-language/)
- [next → FACTS: Table Summarization via Offline Template Generation with Agentic Workflows](https://md.research.elamur.ai/free/en/papers/yuan-2026-facts-table-summarization-offline-template/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/33/)

## References

29 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Walid Ahmad, Elana Simon, Seyone Chithrananda, Gabriel Grand, and Bharath Ramsundar. 2022. Chemberta-2: Towards chemical foundation models. arXiv preprint arXiv:2209.01712. — [arXiv:2209.01712](https://arxiv.org/abs/2209.01712) [find](https://scholar.google.com/scholar?q=Walid+Ahmad%2C+Elana+Simon%2C+Seyone+Chithrananda%2C+Gabriel+Grand%2C+and+Bharath+Ramsundar%2E+2022%2E+Chemberta%2D2%3A+Towards+chemical+foundation+models%2E+arXiv+preprint+arXiv%3A2209%2E01712%2E+%E2%80%94+%5BarXiv%3A2209%2E01712%5D%28https%3A)
2. Esben Jannik Bjerrum. 2017. Smiles enumeration as data augmentation for neural network modeling of molecules. arXiv preprint arXiv:1703.07076. — [arXiv:1703.07076](https://arxiv.org/abs/1703.07076) [find](https://scholar.google.com/scholar?q=Esben+Jannik+Bjerrum%2E+2017%2E+Smiles+enumeration+as+data+augmentation+for+neural+network+modeling+of+molecules%2E+arXiv+preprint+arXiv%3A1703%2E07076%2E+%E2%80%94+%5BarXiv%3A1703%2E07076%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1703%2E07076%29)
3. Johan Broberg, Maria Margareta Bånkestad, and Erik Ylipää Hellqvist. 2022. Pre-training transformers for molecular property prediction using reaction prediction. In ICML 2022 2nd AI for Science Workshop. [find](https://scholar.google.com/scholar?q=Johan+Broberg%2C+Maria+Margareta+B%C3%A5nkestad%2C+and+Erik+Ylip%C3%A4%C3%A4+Hellqvist%2E+2022%2E+Pre%2Dtraining+transformers+for+molecular+property+prediction+using+reaction+prediction%2E+In+ICML+2022+2nd+AI+for+Science+Worksh)
4. Seyone Chithrananda, Gabriel Grand, and Bharath Ramsundar. 2020. Chemberta: large-scale self-supervised pretraining for molecular property prediction. arXiv preprint arXiv:2010.09885. [find](https://scholar.google.com/scholar?q=Seyone+Chithrananda%2C+Gabriel+Grand%2C+and+Bharath+Ramsundar%2E+2020%2E+Chemberta%3A+large%2Dscale+self%2Dsupervised+pretraining+for+molecular+property+prediction%2E+arXiv+preprint+arXiv%3A2010%2E09885%2E)
5. Gonzalo Colmenarejo. 2025. Efgs: a complete and accurate implementation of ertl’s functional group detection algorithm in rdkit. Journal of Chemical Information and Modeling, 65(3):1061–1066. [find](https://scholar.google.com/scholar?q=Gonzalo+Colmenarejo%2E+2025%2E+Efgs%3A+a+complete+and+accurate+implementation+of+ertl%E2%80%99s+functional+group+detection+algorithm+in+rdkit%2E+Journal+of+Chemical+Information+and+Modeling%2C+65%283%29%3A1061%E2%80%931066%2E)
6. Suresh Dara, Swetha Dhamercherla, Surender Singh Jadav, CH Madhu Babu, and Mohamed Jawed Ahsan. 2022. Machine learning in drug discovery: a review. Artificial intelligence review, 55(3):1947–1999. [find](https://scholar.google.com/scholar?q=Suresh+Dara%2C+Swetha+Dhamercherla%2C+Surender+Singh+Jadav%2C+CH+Madhu+Babu%2C+and+Mohamed+Jawed+Ahsan%2E+2022%2E+Machine+learning+in+drug+discovery%3A+a+review%2E+Artificial+intelligence+review%2C+55%283%29%3A1947%E2%80%931999%2E)
7. Peter Ertl. 2017. An algorithm to identify functional groups in organic molecules. Journal of cheminformatics, 9(1):36. [find](https://scholar.google.com/scholar?q=Peter+Ertl%2E+2017%2E+An+algorithm+to+identify+functional+groups+in+organic+molecules%2E+Journal+of+cheminformatics%2C+9%281%29%3A36%2E)
8. Xiaohong Ji, Zhen Wang, Zhifeng Gao, Hang Zheng, Linfeng Zhang, Guolin Ke, and 1 others. 2024. Uni-mol2: Exploring molecular pretraining model at scale. arXiv preprint arXiv:2406.14969. — [arXiv:2406.14969](https://arxiv.org/abs/2406.14969) [find](https://scholar.google.com/scholar?q=Xiaohong+Ji%2C+Zhen+Wang%2C+Zhifeng+Gao%2C+Hang+Zheng%2C+Linfeng+Zhang%2C+Guolin+Ke%2C+and+1+others%2E+2024%2E+Uni%2Dmol2%3A+Exploring+molecular+pretraining+model+at+scale%2E+arXiv+preprint+arXiv%3A2406%2E14969%2E+%E2%80%94+%5BarXiv%3A2406%2E)
9. Sunghwan Kim, Jie Chen, Tiejun Cheng, Asta Gindulyte, Jia He, Siqian He, Qingliang Li, Benjamin A Shoemaker, Paul A Thiessen, Bo Yu, and 1 others. 2019. [find](https://scholar.google.com/scholar?q=Sunghwan+Kim%2C+Jie+Chen%2C+Tiejun+Cheng%2C+Asta+Gindulyte%2C+Jia+He%2C+Siqian+He%2C+Qingliang+Li%2C+Benjamin+A+Shoemaker%2C+Paul+A+Thiessen%2C+Bo+Yu%2C+and+1+others%2E+2019%2E)
10. Pubchem 2019 update: improved access to chemical data. Nucleic acids research, 47(D1):D1102–D1109. [find](https://scholar.google.com/scholar?q=Pubchem+2019+update%3A+improved+access+to+chemical+data%2E+Nucleic+acids+research%2C+47%28D1%29%3AD1102%E2%80%93D1109%2E)
11. Miguelangel Leon, Yuriy Perezhohin, Fernando Peres, Aleš Popovič, and Mauro Castelli. 2024. Comparing smiles and selfies tokenization for enhanced chemical language modeling. Scientific Reports, 14(1):25016. [find](https://scholar.google.com/scholar?q=Miguelangel+Leon%2C+Yuriy+Perezhohin%2C+Fernando+Peres%2C+Ale%C5%A1+Popovic%CC%8C%2C+and+Mauro+Castelli%2E+2024%2E+Comparing+smiles+and+selfies+tokenization+for+enhanced+chemical+language+modeling%2E+Scientific+Reports%2C+14%281)
12. Biaoshun Li, Mujie Lin, Tiegen Chen, and Ling Wang. 2023. Fg-bert: a generalized and self-supervised functional group-based molecular representation learning framework for properties prediction. Briefings in Bioinformatics, 24(6):bbad398. [find](https://scholar.google.com/scholar?q=Biaoshun+Li%2C+Mujie+Lin%2C+Tiegen+Chen%2C+and+Ling+Wang%2E+2023%2E+Fg%2Dbert%3A+a+generalized+and+self%2Dsupervised+functional+group%2Dbased+molecular+representation+learning+framework+for+properties+prediction%2E+Brief)
13. Yuquan Li, Chang-Yu Hsieh, Ruiqiang Lu, Xiaoqing Gong, Xiaorui Wang, Pengyong Li, Shuo Liu, Yanan Tian, Dejun Jiang, Jiaxian Yan, and 1 others. 2022. An adaptive graph learning method for automated molecular interactions and properties predictions. nature machine intelligence, 4(7):645–651. [find](https://scholar.google.com/scholar?q=Yuquan+Li%2C+Chang%2DYu+Hsieh%2C+Ruiqiang+Lu%2C+Xiaoqing+Gong%2C+Xiaorui+Wang%2C+Pengyong+Li%2C+Shuo+Liu%2C+Yanan+Tian%2C+Dejun+Jiang%2C+Jiaxian+Yan%2C+and+1+others%2E+2022%2E+An+adaptive+graph+learning+method+for+automated+mo)
14. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, and Veselin Stoyanov. 2019. [Roberta: A robustly optimized bert pretraining approach](https://md.research.elamur.ai/free/en/papers/liu-2019-roberta-robustly-optimized-bert-pretraining/). arXiv preprint arXiv:1907.11692.
15. Fabrizio Mastrolorito, Fulvio Ciriaco, Maria Vittoria Togo, Nicola Gambacorta, Daniela Trisci-uzzi, Cosimo Damiano Altomare, Nicola Amoroso, Francesca Grisoni, and Orazio Nicolotti. 2025. fragsmiles as a chemical string notation for advanced fragment and chirality representation. Communications Chem… [find](https://scholar.google.com/scholar?q=Fabrizio+Mastrolorito%2C+Fulvio+Ciriaco%2C+Maria+Vittoria+Togo%2C+Nicola+Gambacorta%2C+Daniela+Trisci%2Duzzi%2C+Cosimo+Damiano+Altomare%2C+Nicola+Amoroso%2C+Francesca+Grisoni%2C+and+Orazio+Nicolotti%2E+2025%2E+fragsmiles+a)
16. Leland McInnes, John Healy, and James Melville. 2018. [Umap: Uniform manifold approximation and projection for dimension reduction](https://md.research.elamur.ai/free/en/papers/mcinnes-2018-umap-uniform-manifold-approximation-projection/). arXiv preprint arXiv:1802.03426.
17. Tianhao Peng, Yuchen Li, Xuhong Li, Jiang Bian, Zeke Xie, Ning Sui, Shahid Mumtaz, Yanwu Xu, Linghe Kong, and Haoyi Xiong. 2025. Pre-trained molecular language models with random functional group masking. npj Artificial Intelligence, 1(1):28. [find](https://scholar.google.com/scholar?q=Tianhao+Peng%2C+Yuchen+Li%2C+Xuhong+Li%2C+Jiang+Bian%2C+Zeke+Xie%2C+Ning+Sui%2C+Shahid+Mumtaz%2C+Yanwu+Xu%2C+Linghe+Kong%2C+and+Haoyi+Xiong%2E+2025%2E+Pre%2Dtrained+molecular+language+models+with+random+functional+group+mask)
18. RDKit RDKit. 2024. Open-source cheminformatics. DOI: https://doi.org/10.5281/zenodo, 591637. — [doi:10.5281/zenodo](https://doi.org/10.5281/zenodo) [↗](https://doi.org/10.5281/zenodo)
19. Philippe Schwaller, Teodoro Laino, Théophile Gaudin, Peter Bolgar, Christopher A Hunter, Costas Bekas, and Alpha A Lee. 2019. Molecular transformer: a model for uncertainty-calibrated chemical reaction prediction. ACS central science, 5(9):1572–1583. [find](https://scholar.google.com/scholar?q=Philippe+Schwaller%2C+Teodoro+Laino%2C+Th%C3%A9ophile+Gaudin%2C+Peter+Bolgar%2C+Christopher+A+Hunter%2C+Costas+Bekas%2C+and+Alpha+A+Lee%2E+2019%2E+Molecular+transformer%3A+a+model+for+uncertainty%2Dcalibrated+chemical+reactio)
20. Riya Singh, Aryan Amit Barsainyan, Rida Irfan, Connor Joseph Amorin, Stewart He, Tony Davis, Arun Thiagarajan, Shiva Sankaran, Seyone Chithrananda, Walid Ahmad, and 1 others. 2026. Chemberta-3: an open source training framework for chemical foundation models. Digital Discovery. [find](https://scholar.google.com/scholar?q=Riya+Singh%2C+Aryan+Amit+Barsainyan%2C+Rida+Irfan%2C+Connor+Joseph+Amorin%2C+Stewart+He%2C+Tony+Davis%2C+Arun+Thiagarajan%2C+Shiva+Sankaran%2C+Seyone+Chithrananda%2C+Walid+Ahmad%2C+and+1+others%2E+2026%2E+Chemberta%2D3%3A+an+ope)
21. Young-Han Son, Dong-Hee Shin, and Tae-Eui Kam. 2024. Ftmmr: Fusion transformer for integrating multiple molecular representations. IEEE Journal of Biomedical and Health Informatics, 28(7):4361– 4372. [find](https://scholar.google.com/scholar?q=Young%2DHan+Son%2C+Dong%2DHee+Shin%2C+and+Tae%2DEui+Kam%2E+2024%2E+Ftmmr%3A+Fusion+transformer+for+integrating+multiple+molecular+representations%2E+IEEE+Journal+of+Biomedical+and+Health+Informatics%2C+28%287%29%3A4361%E2%80%93+4372%2E)
22. Mukund Sundararajan, Ankur Taly, and Qiqi Yan. 2017. Axiomatic attribution for deep networks. In International conference on machine learning, pages 3319– 3328. PMLR. [find](https://scholar.google.com/scholar?q=Mukund+Sundararajan%2C+Ankur+Taly%2C+and+Qiqi+Yan%2E+2017%2E+Axiomatic+attribution+for+deep+networks%2E+In+International+conference+on+machine+learning%2C+pages+3319%E2%80%93+3328%2E+PMLR%2E)
23. Sheng Wang, Yuzhi Guo, Yuhong Wang, Hongmao Sun, and Junzhou Huang. 2019. Smiles-bert: large scale unsupervised pre-training for molecular property prediction. In Proceedings of the 10th ACM international conference on bioinformatics, computational biology and health informatics, pages 429–436. [find](https://scholar.google.com/scholar?q=Sheng+Wang%2C+Yuzhi+Guo%2C+Yuhong+Wang%2C+Hongmao+Sun%2C+and+Junzhou+Huang%2E+2019%2E+Smiles%2Dbert%3A+large+scale+unsupervised+pre%2Dtraining+for+molecular+property+prediction%2E+In+Proceedings+of+the+10th+ACM+internati)
24. David Weininger. 1988. Smiles, a chemical language and information system. 1. introduction to methodology and encoding rules. Journal of chemical information and computer sciences, 28(1):31–36. [find](https://scholar.google.com/scholar?q=David+Weininger%2E+1988%2E+Smiles%2C+a+chemical+language+and+information+system%2E+1%2E+introduction+to+methodology+and+encoding+rules%2E+Journal+of+chemical+information+and+computer+sciences%2C+28%281%29%3A31%E2%80%9336%2E)
25. Oliver Wieder, Stefan Kohlbacher, Mélaine Kuenemann, Arthur Garon, Pierre Ducrot, Thomas Seidel, and Thierry Langer. 2020. A compact review of molecular property prediction with graph neural networks. Drug Discovery Today: Technologies, 37:1–12. [find](https://scholar.google.com/scholar?q=Oliver+Wieder%2C+Stefan+Kohlbacher%2C+M%C3%A9laine+Kuenemann%2C+Arthur+Garon%2C+Pierre+Ducrot%2C+Thomas+Seidel%2C+and+Thierry+Langer%2E+2020%2E+A+compact+review+of+molecular+property+prediction+with+graph+neural+networks%2E)
26. Juan-Ni Wu, Tong Wang, Yue Chen, Li-Juan Tang, Hai-Long Wu, and Ru-Qin Yu. 2024. t-smiles: a fragment-based molecular representation framework for de novo ligand design. Nature Communications, 15(1):4993. [find](https://scholar.google.com/scholar?q=Juan%2DNi+Wu%2C+Tong+Wang%2C+Yue+Chen%2C+Li%2DJuan+Tang%2C+Hai%2DLong+Wu%2C+and+Ru%2DQin+Yu%2E+2024%2E+t%2Dsmiles%3A+a+fragment%2Dbased+molecular+representation+framework+for+de+novo+ligand+design%2E+Nature+Communications%2C+15%281%29%3A4)
27. Zhenqin Wu, Bharath Ramsundar, Evan N Feinberg, Joseph Gomes, Caleb Geniesse, Aneesh S Pappu, Karl Leswing, and Vijay Pande. 2018. Moleculenet: a benchmark for molecular machine learning. Chemical science, 9(2):513–530. [find](https://scholar.google.com/scholar?q=Zhenqin+Wu%2C+Bharath+Ramsundar%2C+Evan+N+Feinberg%2C+Joseph+Gomes%2C+Caleb+Geniesse%2C+Aneesh+S+Pappu%2C+Karl+Leswing%2C+and+Vijay+Pande%2E+2018%2E+Moleculenet%3A+a+benchmark+for+molecular+machine+learning%2E+Chemical+sci)
28. Gengmo Zhou, Zhifeng Gao, Qiankun Ding, Hang Zheng, Hongteng Xu, Zhewei Wei, Linfeng Zhang, and Guolin Ke. 2023. Uni-mol: A universal 3d molecular representation learning framework. [find](https://scholar.google.com/scholar?q=Gengmo+Zhou%2C+Zhifeng+Gao%2C+Qiankun+Ding%2C+Hang+Zheng%2C+Hongteng+Xu%2C+Zhewei+Wei%2C+Linfeng+Zhang%2C+and+Guolin+Ke%2E+2023%2E+Uni%2Dmol%3A+A+universal+3d+molecular+representation+learning+framework%2E)
29. Jerret Ross, Brian Belgodere, Vijil Chenthamarakshan, Inkit Padhi, Youssef Mroueh, and Payel Das. 2022. Large-scale chemical language representations capture molecular structure and properties. Nature Machine Intelligence, 4(12):1256–1264. [find](https://scholar.google.com/scholar?q=Jerret+Ross%2C+Brian+Belgodere%2C+Vijil+Chenthamarakshan%2C+Inkit+Padhi%2C+Youssef+Mroueh%2C+and+Payel+Das%2E+2022%2E+Large%2Dscale+chemical+language+representations+capture+molecular+structure+and+properties%2E+Nature)

*Updated 1 November 2026*