---
title: "MCLE-Mol — paper by Pu et al. (2026) — summary and ideas"
description: "Large language models could make molecular property prediction (MPP) explainable. 40 references."
authors: ["Zhili Pu", "Lantian Zhang", "Hao Duan", "Zhixing Zhang", "Keyun Zhu", "Yongqi Fan", "Ruihui Hou", "Tong Ruan", "Yun Tang"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.262/"
doi: "10.18653/v1/2026.findings-acl.262"
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/pu-2026-mcle-mol-empowering-llm-molecular/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/pu-2026-mcle-mol-empowering-llm-molecular/)

[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/36/) / MCLE-Mol: Empowering LLM with Molecular Comprehension and Low-Cost Continual Evolution for Interpretable Property Prediction

# MCLE-Mol: Empowering LLM with Molecular Comprehension and Low-Cost Continual Evolution for Interpretable Property Prediction

**Authors:** Zhili Pu, Lantian Zhang, Hao Duan, Zhixing Zhang, Keyun Zhu, Yongqi Fan, Ruihui Hou, [Tong Ruan](https://md.research.elamur.ai/free/en/authors/tong-ruan/) and [Yun Tang](https://md.research.elamur.ai/free/en/authors/yun-tang/) · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/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.findings-acl.262/) · [PDF](https://aclanthology.org/2026.findings-acl.262.pdf) · [DOI 10.18653/v1/2026.findings-acl.262](https://doi.org/10.18653/v1/2026.findings-acl.262) · [OpenAlex](https://openalex.org/W7166806817)

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

## Summary
Large language models could make molecular property prediction (MPP) explainable. However, they have trouble connecting molecular structure to activity. Earlier methods that supply structural hints also tend to treat chance co-occurrences in training data as chemical laws. MCLE-Mol addresses this with a framework that combines a machine-learning model, an LLM and a rule repository. An RGCN estimates each substructure's contribution to a property using the SME masking technique. These scores are passed to a fine-tuned LLM together with textual descriptions of the molecule at the 1D, 2D and 3D levels, which GPT-4o writes from RDKit-computed descriptors. Context-Calibrated Substructure Attribution Rules (CCSAR) are retrieved for each molecule and correct the raw scores according to its specific chemical context. Over successive iterations the repository gains new rules while the model weights stay frozen. This gives inexpensive adaptation without retraining or catastrophic forgetting. On BACE, BBBP, HIV and Tox21, the authors report that MCLE-Mol beats all baselines and reaches the level of task-specific GNNs while still producing explanations. This makes it relevant for interpretable drug-discovery screening.

## Key points
- **Pipeline:** RGCN with Substructure-Mask Explanation (SME) produces attribution values. GPT-4o generates descriptions anchored on RDKit properties such as LogP, TPSA, NPR1 and NPR2. The top-k CCSAR rules are retrieved, with k = 5 as the default.
- **Ablation:** Performance rises with each added component, from ML attribution to CCSAR to iterative evolution, and the full model is best on all datasets.
- **Rule usage on test sets:** Retrieved rules were used during inference for 98.7% of BACE, 91.9% of BBBP, 90.4% of Tox21 and 88.5% of HIV test molecules.
- **Retrieval depth with k = 5:** ROC-AUC was 0.9178 on BACE and 0.8581 on BBBP. Using k = 7 or k = 9 slightly lowered scores and increased input length.
- **Transfer and evolution:** CCSAR also improved DeepSeek-V3 and Qwen2.5-32B when used as a plug-in. Over three iterations, the number of new rules fell and performance levelled off.

*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

- [East China University of Science and Technology](https://md.research.elamur.ai/free/en/universities/east-china-university-of-science-and-technology/) — University, Shanghai, China · 10 works · 87 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🧮 Computational Theory and Mathematics](https://md.research.elamur.ai/free/en/shelves/computer-science/computational-theory-and-mathematics/) · 252 works · [🧱 Materials Science](https://md.research.elamur.ai/free/en/shelves/materials-science/) · 26 works · [🧱 Materials Chemistry](https://md.research.elamur.ai/free/en/shelves/materials-science/materials-chemistry/) · 26 works · [🧪 Chemistry](https://md.research.elamur.ai/free/en/shelves/chemistry/) · 3 works · [🧪 Physical and Theoretical Chemistry](https://md.research.elamur.ai/free/en/shelves/chemistry/physical-and-theoretical-chemistry/) · 3 works

🧭 Topics (1): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works

## Cite

Cite as: Zhili Pu, Lantian Zhang, Hao Duan, Zhixing Zhang, Keyun Zhu, Yongqi Fan, Ruihui Hou, Tong Ruan and Yun Tang. 2026. MCLE-Mol: Empowering LLM with Molecular Comprehension and Low-Cost Continual Evolution for Interpretable Property Prediction. Findings of the Association for Computational Linguistics: ACL 2026.

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

```bibtex
@inproceedings{pu2026mcle,
  title = {{MCLE-Mol: Empowering LLM with Molecular Comprehension and Low-Cost Continual Evolution for Interpretable Property Prediction}},
  author = {Zhili Pu, Lantian Zhang, Hao Duan, Zhixing Zhang, Keyun Zhu, Yongqi Fan, Ruihui Hou, Tong Ruan and Yun Tang},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.262},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.262}
}
```

- [← previous mcdok at SemEval-2026 Task 13: Finetuning LLMs for Detection of Machine-Generated Code](https://md.research.elamur.ai/free/en/papers/skurla-2026-mcdok-semeval-2026-task-13/)
- [next → mCS-LM: Multimodal Customer Service and Incident Management Systems Based on Large Language Models](https://md.research.elamur.ai/free/en/papers/morales-2026-mcs-lm-multimodal-customer-service/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/36/)

## References

40 references, 10 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Marta Bon, Alan Bilsland, Justin Bower, and Kirsten McAulay. 2022. Fragment-based drug discovery—the importance of high-quality molecule libraries. Molecular Oncology, 16(21):3761–3777. [find](https://scholar.google.com/scholar?q=Marta+Bon%2C+Alan+Bilsland%2C+Justin+Bower%2C+and+Kirsten+McAulay%2E+2022%2E+Fragment%2Dbased+drug+discovery%E2%80%94the+importance+of+high%2Dquality+molecule+libraries%2E+Molecular+Oncology%2C+16%2821%29%3A3761%E2%80%933777%2E)
2. Yuzheng Cai, Siqi Cai, Yuchen Shi, Zihan Xu, Lichao Chen, Yulei Qin, Xiaoyu Tan, Gang Li, Zongyi Li, Haojia Lin, and 1 others. 2025a. Training-free group relative policy optimization. arXiv preprint arXiv:2510.08191. — [arXiv:2510.08191](https://arxiv.org/abs/2510.08191) [find](https://scholar.google.com/scholar?q=Yuzheng+Cai%2C+Siqi+Cai%2C+Yuchen+Shi%2C+Zihan+Xu%2C+Lichao+Chen%2C+Yulei+Qin%2C+Xiaoyu+Tan%2C+Gang+Li%2C+Zongyi+Li%2C+Haojia+Lin%2C+and+1+others%2E+2025a%2E+Training%2Dfree+group+relative+policy+optimization%2E+arXiv+preprint+a)
3. Zhicheng Cai, Xinyuan Guo, Yu Pei, Jiangtao Feng, Jinsong Su, Jiangjie Chen, Ya-Qin Zhang, Wei-Ying Ma, Mingxuan Wang, and Hao Zhou. 2025b. Flex: Continuous agent evolution via forward learning from experience. arXiv preprint arXiv:2511.06449. — [arXiv:2511.06449](https://arxiv.org/abs/2511.06449) [find](https://scholar.google.com/scholar?q=Zhicheng+Cai%2C+Xinyuan+Guo%2C+Yu+Pei%2C+Jiangtao+Feng%2C+Jinsong+Su%2C+Jiangjie+Chen%2C+Ya%2DQin+Zhang%2C+Wei%2DYing+Ma%2C+Mingxuan+Wang%2C+and+Hao+Zhou%2E+2025b%2E+Flex%3A+Continuous+agent+evolution+via+forward+learning+from+e)
4. He Cao, Zijing Liu, Xingyu Lu, Yuan Yao, and Yu Li. 2025. [Instructmol: Multi-modal integration for building a versatile and reliable molecular assistant in drug discovery](https://md.research.elamur.ai/free/en/papers/cao-2025-instructmol-multi-modal-integration-building/). In Proceedings of the 31st International Conference on Computational Linguistics, pages 354– 379.
5. Hongming Chen, Ola Engkvist, Yinhai Wang, Marcus Olivecrona, and Thomas Blaschke. 2018. The rise of deep learning in drug discovery. Drug discovery today, 23(6):1241–1250. [find](https://scholar.google.com/scholar?q=Hongming+Chen%2C+Ola+Engkvist%2C+Yinhai+Wang%2C+Marcus+Olivecrona%2C+and+Thomas+Blaschke%2E+2018%2E+The+rise+of+deep+learning+in+drug+discovery%2E+Drug+discovery+today%2C+23%286%29%3A1241%E2%80%931250%2E)
6. Artem Cherkasov, Eugene N Muratov, Denis Fourches, Alexandre Varnek, Igor I Baskin, Mark Cronin, John Dearden, Paola Gramatica, Yvonne C Martin, Roberto Todeschini, and 1 others. 2014. Qsar modeling: where have you been? where are you going to? Journal of medicinal chemistry, 57(12):4977–5010. [find](https://scholar.google.com/scholar?q=Artem+Cherkasov%2C+Eugene+N+Muratov%2C+Denis+Fourches%2C+Alexandre+Varnek%2C+Igor+I+Baskin%2C+Mark+Cronin%2C+John+Dearden%2C+Paola+Gramatica%2C+Yvonne+C+Martin%2C+Roberto+Todeschini%2C+and+1+others%2E+2014%2E+Qsar+modeling%3A+)
7. Carl Edwards, Tuan Lai, Kevin Ros, Garrett Honke, Kyunghyun Cho, and Heng Ji. 2022. [Translation between molecules and natural language](https://md.research.elamur.ai/free/en/papers/edwards-2022-translation-between-molecules-natural-language/). arXiv preprint arXiv:2204.11817.
8. Xiaomin Fang, Lihang Liu, Jieqiong Lei, Donglong He, Shanzhuo Zhang, Jingbo Zhou, Fan Wang, Hua Wu, and Haifeng Wang. 2022. Geometry-enhanced molecular representation learning for property prediction. Nature Machine Intelligence, 4(2):127–134. [find](https://scholar.google.com/scholar?q=Xiaomin+Fang%2C+Lihang+Liu%2C+Jieqiong+Lei%2C+Donglong+He%2C+Shanzhuo+Zhang%2C+Jingbo+Zhou%2C+Fan+Wang%2C+Hua+Wu%2C+and+Haifeng+Wang%2E+2022%2E+Geometry%2Denhanced+molecular+representation+learning+for+property+prediction%2E)
9. Luyu Gao, Xueguang Ma, Jimmy Lin, and Jamie Callan. 2023. [Precise zero-shot dense retrieval without relevance labels](https://md.research.elamur.ai/free/en/papers/gao-2022-precise-zero-shot-dense-retrieval/). In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1762–1777.
10. Justin Gilmer, Samuel S Schoenholz, Patrick F Riley, Oriol Vinyals, and George E Dahl. 2017. [Neural message passing for quantum chemistry](https://md.research.elamur.ai/free/en/papers/gilmer-2017-neural-message-passing-quantum-chemistry/). In International conference on machine learning, pages 1263–1272. Pmlr.
11. Taicheng Guo, Kehan Guo, Bozhao Nan, Zhenwen Liang, Zhichun Guo, Nitesh V Chawla, Olaf Wiest, and Xiangliang Zhang. What can large language models do in chemistry? a comprehensive benchmark on eight tasks, 2023. URL https://arxiv.org/abs/2305.18365. — [arXiv:2305.18365](https://arxiv.org/abs/2305.18365) [↗](https://arxiv.org/abs/2305.18365)
12. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, Weizhu Chen, and 1 others. 2022. [Lora: Low-rank adaptation of large language models](https://md.research.elamur.ai/free/en/papers/hu-2021-lora-low-rank-adaptation-large/). ICLR, 1(2):3.
13. Aaron Hurst, Adam Lerer, Adam P Goucher, Adam Perelman, Aditya Ramesh, Aidan Clark, AJ Ostrow, Akila Welihinda, Alan Hayes, Alec Radford, and 1 others. 2024. [Gpt-4o system card](https://md.research.elamur.ai/free/en/papers/openai-2024-gpt-4o-system-card/). arXiv preprint arXiv:2410.21276.
14. José Jiménez-Luna, Francesca Grisoni, and Gisbert Schneider. 2020. Drug discovery with explainable artificial intelligence. Nature Machine Intelligence, 2(10):573–584. [find](https://scholar.google.com/scholar?q=Jos%C3%A9+Jim%C3%A9nez%2DLuna%2C+Francesca+Grisoni%2C+and+Gisbert+Schneider%2E+2020%2E+Drug+discovery+with+explainable+artificial+intelligence%2E+Nature+Machine+Intelligence%2C+2%2810%29%3A573%E2%80%93584%2E)
15. Jiaxin Ju, Yizhen Zheng, Huan Yee Koh, Can Wang, and Shirui Pan. 2025. M2llm: Multi-view molecular representation learning with large language models. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, pages 7437– 7445. [find](https://scholar.google.com/scholar?q=Jiaxin+Ju%2C+Yizhen+Zheng%2C+Huan+Yee+Koh%2C+Can+Wang%2C+and+Shirui+Pan%2E+2025%2E+M2llm%3A+Multi%2Dview+molecular+representation+learning+with+large+language+models%2E+In+Proceedings+of+the+Thirty%2DFourth+International)
16. Nikhil Kandpal, Haikang Deng, Adam Roberts, Eric Wallace, and Colin Raffel. 2023. [Large language models struggle to learn long-tail knowledge](https://md.research.elamur.ai/free/en/papers/kandpal-2022-large-language-models-struggle-learn/). In International conference on machine learning, pages 15696–15707. PMLR.
17. James Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, and 1 others. 2017. Overcoming catastrophic forgetting in neural networks. Proceedings of the national academy of sciences, 114(1… [find](https://scholar.google.com/scholar?q=James+Kirkpatrick%2C+Razvan+Pascanu%2C+Neil+Rabinowitz%2C+Joel+Veness%2C+Guillaume+Desjardins%2C+Andrei+A+Rusu%2C+Kieran+Milan%2C+John+Quan%2C+Tiago+Ramalho%2C+Agnieszka+Grabska%2DBarwinska%2C+and+1+others%2E+2017%2E+Overcomin)
18. Greg Landrum. 2006. Rdkit: Open-source cheminformatics. Accessed: 2026-01-04. [find](https://scholar.google.com/scholar?q=Greg+Landrum%2E+2006%2E+Rdkit%3A+Open%2Dsource+cheminformatics%2E+Accessed%3A+2026%2D01%2D04%2E)
19. Chanhui Lee, Hanbum Ko, Yuheon Song, YongJun Jeong, Rodrigo Hormazabal, Sehui Han, Kyunghoon Bae, Sungbin Lim, and Sungwoong Kim. 2025. Mol-llm: Multimodal generalist molecular llm with improved graph utilization. arXiv preprint arXiv:2502.02810. — [arXiv:2502.02810](https://arxiv.org/abs/2502.02810) [find](https://scholar.google.com/scholar?q=Chanhui+Lee%2C+Hanbum+Ko%2C+Yuheon+Song%2C+YongJun+Jeong%2C+Rodrigo+Hormazabal%2C+Sehui+Han%2C+Kyunghoon+Bae%2C+Sungbin+Lim%2C+and+Sungwoong+Kim%2E+2025%2E+Mol%2Dllm%3A+Multimodal+generalist+molecular+llm+with+improved+graph)
20. Bangjiang Lin, Shujie Yan, and Bowen Zhen. 2025. A machine learning method for predicting molecular antimicrobial activity. Scientific Reports, 15(1):6559. [find](https://scholar.google.com/scholar?q=Bangjiang+Lin%2C+Shujie+Yan%2C+and+Bowen+Zhen%2E+2025%2E+A+machine+learning+method+for+predicting+molecular+antimicrobial+activity%2E+Scientific+Reports%2C+15%281%29%3A6559%2E)
21. Pengfei Liu, Yiming Ren, Jun Tao, and Zhixiang Ren. 2024a. Git-mol: A multi-modal large language model for molecular science with graph, image, and text. Computers in biology and medicine, 171:108073. [find](https://scholar.google.com/scholar?q=Pengfei+Liu%2C+Yiming+Ren%2C+Jun+Tao%2C+and+Zhixiang+Ren%2E+2024a%2E+Git%2Dmol%3A+A+multi%2Dmodal+large+language+model+for+molecular+science+with+graph%2C+image%2C+and+text%2E+Computers+in+biology+and+medicine%2C+171%3A108073%2E)
22. Yuyan Liu, Sirui Ding, Sheng Zhou, Wenqi Fan, and Qiaoyu Tan. 2024b. Moleculargpt: Open large language model (llm) for few-shot molecular property prediction. arXiv preprint arXiv:2406.12950. — [arXiv:2406.12950](https://arxiv.org/abs/2406.12950) [find](https://scholar.google.com/scholar?q=Yuyan+Liu%2C+Sirui+Ding%2C+Sheng+Zhou%2C+Wenqi+Fan%2C+and+Qiaoyu+Tan%2E+2024b%2E+Moleculargpt%3A+Open+large+language+model+%28llm%29+for+few%2Dshot+molecular+property+prediction%2E+arXiv+preprint+arXiv%3A2406%2E12950%2E+%E2%80%94+%5BarXiv)
23. Zequn Liu, Wei Zhang, Yingce Xia, Lijun Wu, Shufang Xie, Tao Qin, Ming Zhang, and Tie-Yan Liu. 2023a. Molxpt: Wrapping molecules with text for generative pre-training. arXiv preprint arXiv:2305.10688. [find](https://scholar.google.com/scholar?q=Zequn+Liu%2C+Wei+Zhang%2C+Yingce+Xia%2C+Lijun+Wu%2C+Shufang+Xie%2C+Tao+Qin%2C+Ming+Zhang%2C+and+Tie%2DYan+Liu%2E+2023a%2E+Molxpt%3A+Wrapping+molecules+with+text+for+generative+pre%2Dtraining%2E+arXiv+preprint+arXiv%3A2305%2E10688%2E)
24. Zhiyuan Liu, Sihang Li, Yanchen Luo, Hao Fei, Yixin Cao, Kenji Kawaguchi, Xiang Wang, and Tat-Seng Chua. 2023b. Molca: Molecular graph-language modeling with cross-modal projector and uni-modal adapter. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages… [find](https://scholar.google.com/scholar?q=Zhiyuan+Liu%2C+Sihang+Li%2C+Yanchen+Luo%2C+Hao+Fei%2C+Yixin+Cao%2C+Kenji+Kawaguchi%2C+Xiang+Wang%2C+and+Tat%2DSeng+Chua%2E+2023b%2E+Molca%3A+Molecular+graph%2Dlanguage+modeling+with+cross%2Dmodal+projector+and+uni%2Dmodal+adapte)
25. Shayne Longpre, Kartik Perisetla, Anthony Chen, Nikhil Ramesh, Chris DuBois, and Sameer Singh. 2021. [Entity-based knowledge conflicts in question answering](https://md.research.elamur.ai/free/en/papers/longpre-2021-entity-based-knowledge-conflicts-question/). arXiv preprint arXiv:2109.05052.
26. ModelScope. 2023. Swift: Scalable lightweight infrastructure for fine-tuning. [https://github.com/](https://github.com/) modelscope/swift. [↗](https://github.com/)
27. David Rogers and Mathew Hahn. 2010. Extended-connectivity fingerprints. Journal of chemical information and modeling, 50(5):742–754. [find](https://scholar.google.com/scholar?q=David+Rogers+and+Mathew+Hahn%2E+2010%2E+Extended%2Dconnectivity+fingerprints%2E+Journal+of+chemical+information+and+modeling%2C+50%285%29%3A742%E2%80%93754%2E)
28. Dagmar Stumpfe and Jurgen Bajorath. 2012. Exploring activity cliffs in medicinal chemistry: miniperspec-tive. Journal of medicinal chemistry, 55(7):2932– 2942. [find](https://scholar.google.com/scholar?q=Dagmar+Stumpfe+and+Jurgen+Bajorath%2E+2012%2E+Exploring+activity+cliffs+in+medicinal+chemistry%3A+miniperspec%2Dtive%2E+Journal+of+medicinal+chemistry%2C+55%287%29%3A2932%E2%80%93+2942%2E)
29. Jessica Vamathevan, Dominic Clark, Paul Czodrowski, Ian Dunham, Edgardo Ferran, George Lee, Bin Li, Anant Madabhushi, Parantu Shah, Michaela Spitzer, and 1 others. 2019. Applications of machine learning in drug discovery and development. Nature reviews Drug discovery, 18(6):463–477. [find](https://scholar.google.com/scholar?q=Jessica+Vamathevan%2C+Dominic+Clark%2C+Paul+Czodrowski%2C+Ian+Dunham%2C+Edgardo+Ferran%2C+George+Lee%2C+Bin+Li%2C+Anant+Madabhushi%2C+Parantu+Shah%2C+Michaela+Spitzer%2C+and+1+others%2E+2019%2E+Applications+of+machine+learni)
30. 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)
31. Zhenxing Wu, Jike Wang, Hongyan Du, Dejun Jiang, Yu Kang, Dan Li, Peichen Pan, Yafeng Deng, Dongsheng Cao, Chang-Yu Hsieh, and 1 others. 2023. Chemistry-intuitive explanation of graph neural networks for molecular property prediction with substructure masking. Nature Communications, 14(1):2585. [find](https://scholar.google.com/scholar?q=Zhenxing+Wu%2C+Jike+Wang%2C+Hongyan+Du%2C+Dejun+Jiang%2C+Yu+Kang%2C+Dan+Li%2C+Peichen+Pan%2C+Yafeng+Deng%2C+Dongsheng+Cao%2C+Chang%2DYu+Hsieh%2C+and+1+others%2E+2023%2E+Chemistry%2Dintuitive+explanation+of+graph+neural+networks+)
32. Ziting Xian, Jiawei Gu, Lingbo Li, and Shangsong Liang. 2025. Molrag: unlocking the power of large language models for molecular property prediction. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 15513–15531. [find](https://scholar.google.com/scholar?q=Ziting+Xian%2C+Jiawei+Gu%2C+Lingbo+Li%2C+and+Shangsong+Liang%2E+2025%2E+Molrag%3A+unlocking+the+power+of+large+language+models+for+molecular+property+prediction%2E+In+Proceedings+of+the+63rd+Annual+Meeting+of+the+A)
33. Shitao Xiao, Zheng Liu, Peitian Zhang, Niklas Muennighoff, Defu Lian, and Jian-Yun Nie. 2024. [C-pack: Packed resources for general chinese embeddings](https://md.research.elamur.ai/free/en/papers/xiao-2023-c-pack-packed-resources/). In Proceedings of the 47th international ACM SIGIR conference on research and development in information retrieval, pages 641–649.
34. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. 2025. [Qwen3 technical report](https://md.research.elamur.ai/free/en/papers/yang-2025-qwen3-technical-report/). arXiv preprint arXiv:2505.09388.
35. Kevin Yang, Kyle Swanson, Wengong Jin, Connor Coley, Philipp Eiden, Hua Gao, Angel Guzman-Perez, Timothy Hopper, Brian Kelley, Miriam Mathea, and 1 others. 2019. Analyzing learned molecular representations for property prediction. Journal of chemical information and modeling, 59(8):3370–3388. [find](https://scholar.google.com/scholar?q=Kevin+Yang%2C+Kyle+Swanson%2C+Wengong+Jin%2C+Connor+Coley%2C+Philipp+Eiden%2C+Hua+Gao%2C+Angel+Guzman%2DPerez%2C+Timothy+Hopper%2C+Brian+Kelley%2C+Miriam+Mathea%2C+and+1+others%2E+2019%2E+Analyzing+learned+molecular+representa)
36. Chengxuan Ying, Tianle Cai, Shengjie Luo, Shuxin Zheng, Guolin Ke, Di He, Yanming Shen, and Tie-Yan Liu. 2021. Do transformers really perform badly for graph representation? Advances in neural information processing systems, 34:28877–28888. [find](https://scholar.google.com/scholar?q=Chengxuan+Ying%2C+Tianle+Cai%2C+Shengjie+Luo%2C+Shuxin+Zheng%2C+Guolin+Ke%2C+Di+He%2C+Yanming+Shen%2C+and+Tie%2DYan+Liu%2E+2021%2E+Do+transformers+really+perform+badly+for+graph+representation%3F+Advances+in+neural+informa)
37. Yizhen Zheng, Huan Yee Koh, Jiaxin Ju, Can Wang, and Shirui Pan. 2025. Large language models for scientific discovery in molecular property prediction. Nature Machine Intelligence. [find](https://scholar.google.com/scholar?q=Yizhen+Zheng%2C+Huan+Yee+Koh%2C+Jiaxin+Ju%2C+Can+Wang%2C+and+Shirui+Pan%2E+2025%2E+Large+language+models+for+scientific+discovery+in+molecular+property+prediction%2E+Nature+Machine+Intelligence%2E)
38. Chunting Zhou, Pengfei Liu, Puxin Xu, Srinivasan Iyer, Jiao Sun, Yuning Mao, Xuezhe Ma, Avia Efrat, Ping Yu, Lili Yu, and 1 others. 2023a. Lima: Less is more for alignment. Advances in Neural Information Processing Systems, 36:55006–55021. [find](https://scholar.google.com/scholar?q=Chunting+Zhou%2C+Pengfei+Liu%2C+Puxin+Xu%2C+Srinivasan+Iyer%2C+Jiao+Sun%2C+Yuning+Mao%2C+Xuezhe+Ma%2C+Avia+Efrat%2C+Ping+Yu%2C+Lili+Yu%2C+and+1+others%2E+2023a%2E+Lima%3A+Less+is+more+for+alignment%2E+Advances+in+Neural+Informat)
39. Gengmo Zhou, Zhifeng Gao, Qiankun Ding, Hang Zheng, Hongteng Xu, Zhewei Wei, Linfeng Zhang, and Guolin Ke. 2023b. 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+2023b%2E+Uni%2Dmol%3A+A+universal+3d+molecular+representation+learning+framework%2E)
40. Jinhua Zhu, Yingce Xia, Lijun Wu, Shufang Xie, Wengang Zhou, Tao Qin, Houqiang Li, and Tie-Yan Liu. 2023. Dual-view molecular pre-training. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 3615– 3627. [find](https://scholar.google.com/scholar?q=Jinhua+Zhu%2C+Yingce+Xia%2C+Lijun+Wu%2C+Shufang+Xie%2C+Wengang+Zhou%2C+Tao+Qin%2C+Houqiang+Li%2C+and+Tie%2DYan+Liu%2E+2023%2E+Dual%2Dview+molecular+pre%2Dtraining%2E+In+Proceedings+of+the+29th+ACM+SIGKDD+Conference+on+Knowledg)

*Updated 1 November 2026*