---
title: "ChemVLR — paper by Zhao et al. (2026) — summary and ideas"
description: "The work offers a reproducible recipe for interpretable multimodal chemistry assistants. 48 references."
authors: ["Xuanle Zhao", "Xinyuan Cai", "Xiang Cheng", "Xiuyi Chen", "Bo XU"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.1206/"
doi: "10.18653/v1/2026.findings-acl.1206"
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/zhao-2026-chemvlr-prioritizing-reasoning-perception-chemical/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/zhao-2026-chemvlr-prioritizing-reasoning-perception-chemical/)

[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/26/) / ChemVLR: Prioritizing Reasoning in Perception for Chemical Vision-Language Understanding

# ChemVLR: Prioritizing Reasoning in Perception for Chemical Vision-Language Understanding

**Authors:** [Xuanle Zhao](https://md.research.elamur.ai/free/en/authors/xuanle-zhao/), [Xinyuan Cai](https://md.research.elamur.ai/free/en/authors/xinyuan-cai/), Xiang Cheng, [Xiuyi Chen](https://md.research.elamur.ai/free/en/authors/xiuyi-chen/) and [Bo XU](https://md.research.elamur.ai/free/en/authors/bo-xu/) · **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.1206/) · [PDF](https://aclanthology.org/2026.findings-acl.1206.pdf) · [DOI 10.18653/v1/2026.findings-acl.1206](https://doi.org/10.18653/v1/2026.findings-acl.1206) · [OpenAlex](https://openalex.org/W7166805986)

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

## Summary
ChemVLR addresses a gap in chemical vision-language modeling. Existing chemical VLMs such as ChemVLM and TinyChemVL map images of molecules or reactions directly to answers. They give no interpretable rationale, and they make little use of the LLM's latent chemical knowledge. ChemVLR instead reasons explicitly during perception: it first lists fine-grained descriptors such as functional groups and reaction sites, then commits to a SMILES answer.

To get training data, the authors used "cross-modality reverse-engineering." Textual SMILES question–answer pairs from ORDerly were given to Gemini-2.5-Flash, which wrote plausible visual reasoning traces. These were anchored by IUPAC names, RDKit-derived functional groups and hand-written expert demonstrations. The outputs passed a three-step filter and were rendered as images with RDKit and Indigo.

Training on Qwen2.5/3-VL backbones has three stages: continual pre-training on captions, SFT on reasoning plus instruction data, and DAPO-based RL. The resulting model reportedly beats proprietary and chemistry-specific baselines. The work offers a reproducible recipe for interpretable multimodal chemistry assistants.

## Key points
- **Dataset size and composition:** about 760k curated samples.
  - 400k captions (molecule and reaction).
  - 168k recognition samples (95k molecular, 73k reaction).
  - 192k reaction-prediction samples.
- **Filtering pipeline:** structural checks on reasoning traces, answer-consistency validation, and external review by GPT-4.1-mini.
- **Semantic anchors (Table 1):** adding demos, IUPAC names and RDKit groups raised high-quality retention over SMILES-only prompting.
  - Molecular recognition: from 78% to 95%.
  - Reaction recognition: from 58% to 73%.
  - Reaction prediction: from 55% to 76.8%.
- **RL reward design:** the reward fires only when Tanimoto similarity equals 1.0 (structural identity). It outperformed both a dense Tanimoto reward and exact string matching (Table 6).
  - img2smiles: 92.7 vs. 90.2 (Tanimoto) and 90.8 (exact match).
  - Rxn-V R: 74.4.
  - Rxn-V P: 67.8.
- **Limitations and release:** some generated traces still contain errors. Tasks such as property prediction and K-12-style real-world images are not yet covered. Code and weights are promised at github.com/xxlllz/ChemVLR.

*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

- [Chinese Academy of Sciences](https://md.research.elamur.ai/free/en/government/chinese-academy-of-sciences/) — Government body, Beijing, China · 246 works · 24,951 citations
- [Institute of Automation](https://md.research.elamur.ai/free/en/labs/institute-of-automation/) — Research institute, Beijing, China · 87 works · 14,883 citations
- [University of Chinese Academy of Sciences](https://md.research.elamur.ai/free/en/universities/university-of-chinese-academy-of-sciences/) — University, Beijing, China · 188 works · 8,743 citations

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works · [🧮 Computational Theory and Mathematics](https://md.research.elamur.ai/free/en/shelves/computer-science/computational-theory-and-mathematics/) · 321 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

🧭 Topics (2): [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,685 works · [🖼️ Vision & language](https://md.research.elamur.ai/free/en/topics/multimodal/) · 811 works

## Cite

Cite as: Xuanle Zhao, Xinyuan Cai, Xiang Cheng, Xiuyi Chen and Bo XU. 2026. ChemVLR: Prioritizing Reasoning in Perception for Chemical Vision-Language Understanding. 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{zhao2026chemvlr,
  title = {{ChemVLR: Prioritizing Reasoning in Perception for Chemical Vision-Language Understanding}},
  author = {Xuanle Zhao, Xinyuan Cai, Xiang Cheng, Xiuyi Chen and Bo XU},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.1206},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.1206}
}
```

- [← previous ChemReason-Bench: Benchmarking Large Language Models for Procedural Reasoning in Experimental Chemistry](https://md.research.elamur.ai/free/en/papers/zhang-2026-chemreason-bench-benchmarking-large-language/)
- [next → Child-directed speech facilitates production, not comprehension, in BabyLMs](https://md.research.elamur.ai/free/en/papers/bunzeck-2026-child-directed-speech-facilitates-production/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/26/)

## References

48 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Marah Abdin, Jyoti Aneja, Hany Awadalla, Ahmed Awadallah, Ammar Ahmad Awan, Nguyen Bach, Amit Bahree, Arash Bakhtiari, Jianmin Bao, Harkirat Behl, and 1 others. 2024. [Phi-3 technical report: A highly capable language model locally on your phone](https://md.research.elamur.ai/free/en/papers/abdin-2024-phi-3-technical-report-highly/). arXiv preprint arXiv:2404.14219.
2. Shuai Bai, Yuxuan Cai, Ruizhe Chen, Keqin Chen, Xionghui Chen, Zesen Cheng, Lianghao Deng, Wei Ding, Chang Gao, Chunjiang Ge, Wenbin Ge, Zhifang Guo, Qidong Huang, Jie Huang, Fei Huang, Binyuan Hui, Shutong Jiang, Zhaohai Li, Mingsheng Li, and 45 others. 2025a. [Qwen3-vl technical report](https://md.research.elamur.ai/free/en/papers/bai-2025-qwen3-vl-technical-report/). Preprint, a…
3. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, and 1 others. 2025b. Qwen2. 5-vl technical report. arXiv preprint arXiv:2502.13923. — [Qwen2.5-VL Technical Report](https://md.research.elamur.ai/free/en/papers/bai-2025-qwen2-5-vl-technical/)
4. A Patrícia Bento, Anne Hersey, Eloy Félix, Greg Landrum, Anna Gaulton, Francis Atkinson, Louisa J Bellis, Marleen De Veij, and Andrew R Leach. 2020. An open source chemical structure curation pipeline using rdkit. Journal of Cheminformatics, 12:1–16. [find](https://scholar.google.com/scholar?q=A+Patr%C3%ADcia+Bento%2C+Anne+Hersey%2C+Eloy+F%C3%A9lix%2C+Greg+Landrum%2C+Anna+Gaulton%2C+Francis+Atkinson%2C+Louisa+J+Bellis%2C+Marleen+De+Veij%2C+and+Andrew+R+Leach%2E+2020%2E+An+open+source+chemical+structure+curation+pipeline)
5. Jinhe Bi, Yifan Wang, Danqi Yan, Xun Xiao, Artur Hecker, Volker Tresp, and Yunpu Ma. 2025a. Prism: Self-pruning intrinsic selection method for training-free multimodal data selection. arXiv preprint arXiv:2502.12119. — [arXiv:2502.12119](https://arxiv.org/abs/2502.12119) [find](https://scholar.google.com/scholar?q=Jinhe+Bi%2C+Yifan+Wang%2C+Danqi+Yan%2C+Xun+Xiao%2C+Artur+Hecker%2C+Volker+Tresp%2C+and+Yunpu+Ma%2E+2025a%2E+Prism%3A+Self%2Dpruning+intrinsic+selection+method+for+training%2Dfree+multimodal+data+selection%2E+arXiv+preprint+a)
6. Jinhe Bi, Danqi Yan, Yifan Wang, Wenke Huang, Haokun Chen, Guancheng Wan, Mang Ye, Xun Xiao, Hinrich Schuetze, Volker Tresp, and 1 others. 2025b. Cot-kinetics: A theoretical modeling assessing lrm reasoning process. arXiv preprint arXiv:2505.13408. — [arXiv:2505.13408](https://arxiv.org/abs/2505.13408) [find](https://scholar.google.com/scholar?q=Jinhe+Bi%2C+Danqi+Yan%2C+Yifan+Wang%2C+Wenke+Huang%2C+Haokun+Chen%2C+Guancheng+Wan%2C+Mang+Ye%2C+Xun+Xiao%2C+Hinrich+Schuetze%2C+Volker+Tresp%2C+and+1+others%2E+2025b%2E+Cot%2Dkinetics%3A+A+theoretical+modeling+assessing+lrm+rea)
7. Qiguang Chen, Libo Qin, Jinhao Liu, Dengyun Peng, Jiannan Guan, Peng Wang, Mengkang Hu, Yuhang Zhou, Te Gao, and Wanxiang Che. 2025. [Towards reasoning era: A survey of long chain-of-thought for reasoning large language models](https://md.research.elamur.ai/free/en/papers/chen-2025-towards-reasoning-era-survey-long/). arXiv preprint arXiv:2503.09567.
8. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, and 1 others. 2025. [Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities](https://md.research.elamur.ai/free/en/papers/comanici-2025-gemini-2-5-pushing-frontier/). arX…
9. Hongyuan Dong, Zijian Kang, Weijie Yin, Xiao Liang, Chao Feng, and Jiao Ran. 2025. Scalable vision language model training via high quality data curation. arXiv preprint arXiv:2501.05952. [find](https://scholar.google.com/scholar?q=Hongyuan+Dong%2C+Zijian+Kang%2C+Weijie+Yin%2C+Xiao+Liang%2C+Chao+Feng%2C+and+Jiao+Ran%2E+2025%2E+Scalable+vision+language+model+training+via+high+quality+data+curation%2E+arXiv+preprint+arXiv%3A2501%2E05952%2E)
10. Patrick Esser, Robin Rombach, and Bjorn Ommer. 2021. Taming transformers for high-resolution image synthesis. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 12873–12883. [find](https://scholar.google.com/scholar?q=Patrick+Esser%2C+Robin+Rombach%2C+and+Bjorn+Ommer%2E+2021%2E+Taming+transformers+for+high%2Dresolution+image+synthesis%2E+In+Proceedings+of+the+IEEE%2FCVF+conference+on+computer+vision+and+pattern+recognition%2C+page)
11. Google DeepMind. 2025. Gemini 3 flash. [https://deepmind.google/models/gemini/flash/](https://deepmind.google/models/gemini/flash/). Accessed: 2025-12-28. [↗](https://deepmind.google/models/gemini/flash/)
12. OpenAI. 2025a. Introducing GPT-4.1 in the API. [https://openai.com/index/gpt-4-1/](https://openai.com/index/gpt-4-1/). Accessed: 2025-12-28. [↗](https://openai.com/index/gpt-4-1/)
13. Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, and 1 others. 2025. Deepseekr1 incentivizes reasoning in llms through reinforcement learning. Nature, 645(8081):633–638. [find](https://scholar.google.com/scholar?q=Daya+Guo%2C+Dejian+Yang%2C+Haowei+Zhang%2C+Junxiao+Song%2C+Peiyi+Wang%2C+Qihao+Zhu%2C+Runxin+Xu%2C+Ruoyu+Zhang%2C+Shirong+Ma%2C+Xiao+Bi%2C+and+1+others%2E+2025%2E+Deepseekr1+incentivizes+reasoning+in+llms+through+reinforceme)
14. OpenAI. 2025b. Introducing GPT-5. [https://openai.com/zh-Hans-CN/index/introducing-gpt-5/](https://openai.com/zh-Hans-CN/index/introducing-gpt-5/). Accessed: 2025-12-28. [↗](https://openai.com/zh-Hans-CN/index/introducing-gpt-5/)
15. Yang Han, Ziping Wan, Lu Chen, Kai Yu, and Xin Chen. 2025. From generalist to specialist: A survey of large language models for chemistry. In Proceedings of the 31st International Conference on Computational Linguistics, pages 1106–1123. [find](https://scholar.google.com/scholar?q=Yang+Han%2C+Ziping+Wan%2C+Lu+Chen%2C+Kai+Yu%2C+and+Xin+Chen%2E+2025%2E+From+generalist+to+specialist%3A+A+survey+of+large+language+models+for+chemistry%2E+In+Proceedings+of+the+31st+International+Conference+on+Comput)
16. Wenxuan Huang, Bohan Jia, Zijie Zhai, Shaosheng Cao, Zheyu Ye, Fei Zhao, Zhe Xu, Yao Hu, and Shaohui Lin. 2025. [Vision-r1: Incentivizing reasoning capability in multimodal large language models](https://md.research.elamur.ai/free/en/papers/huang-2025-vision-r1-incentivizing-reasoning-capability/). arXiv preprint arXiv:2503.06749.
17. 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. 2023. [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+2023%2E)
18. Pubchem 2023 update. Nucleic acids research, 51(D1):D1373–D1380. [find](https://scholar.google.com/scholar?q=Pubchem+2023+update%2E+Nucleic+acids+research%2C+51%28D1%29%3AD1373%E2%80%93D1380%2E)
19. Jiatong Li, Weida Wang, Qinggang Zhang, Junxian Li, Di Zhang, Changmeng Zheng, Shufei Zhang, Xiaoyong Wei, and Qing Li. 2025a. Mol-r1: Towards explicit long-cot reasoning in molecule discovery. arXiv preprint arXiv:2508.08401. — [arXiv:2508.08401](https://arxiv.org/abs/2508.08401) [find](https://scholar.google.com/scholar?q=Jiatong+Li%2C+Weida+Wang%2C+Qinggang+Zhang%2C+Junxian+Li%2C+Di+Zhang%2C+Changmeng+Zheng%2C+Shufei+Zhang%2C+Xiaoyong+Wei%2C+and+Qing+Li%2E+2025a%2E+Mol%2Dr1%3A+Towards+explicit+long%2Dcot+reasoning+in+molecule+discovery%2E+arXiv+)
20. Junxian Li, Di Zhang, Xunzhi Wang, Zeying Hao, Jingdi Lei, Qian Tan, Cai Zhou, Wei Liu, Yaotian Yang, Xinrui Xiong, and 1 others. 2025b. Chemvlm: Exploring the power of multimodal large language models in chemistry area. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pa… [find](https://scholar.google.com/scholar?q=Junxian+Li%2C+Di+Zhang%2C+Xunzhi+Wang%2C+Zeying+Hao%2C+Jingdi+Lei%2C+Qian+Tan%2C+Cai+Zhou%2C+Wei+Liu%2C+Yaotian+Yang%2C+Xinrui+Xiong%2C+and+1+others%2E+2025b%2E+Chemvlm%3A+Exploring+the+power+of+multimodal+large+language+model)
21. Pengfei Liu, Jun Tao, and Zhixiang Ren. 2025. A quantitative analysis of knowledge-learning preferences in large language models in molecular science. Nature Machine Intelligence, pages 1–13. [find](https://scholar.google.com/scholar?q=Pengfei+Liu%2C+Jun+Tao%2C+and+Zhixiang+Ren%2E+2025%2E+A+quantitative+analysis+of+knowledge%2Dlearning+preferences+in+large+language+models+in+molecular+science%2E+Nature+Machine+Intelligence%2C+pages+1%E2%80%9313%2E)
22. Gary L Long and James D Winefordner. 1983. Limit of detection. a closer look at the iupac definition. Analytical chemistry, 55(7):712A–724A. [find](https://scholar.google.com/scholar?q=Gary+L+Long+and+James+D+Winefordner%2E+1983%2E+Limit+of+detection%2E+a+closer+look+at+the+iupac+definition%2E+Analytical+chemistry%2C+55%287%29%3A712A%E2%80%93724A%2E)
23. Lucas Morin, Martin Danelljan, Maria Isabel Agea, Ahmed Nassar, Valery Weber, Ingmar Meijer, Peter Staar, and Fisher Yu. 2023. Molgrapher: graph-based visual recognition of chemical structures. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 19552–19561. [find](https://scholar.google.com/scholar?q=Lucas+Morin%2C+Martin+Danelljan%2C+Maria+Isabel+Agea%2C+Ahmed+Nassar%2C+Valery+Weber%2C+Ingmar+Meijer%2C+Peter+Staar%2C+and+Fisher+Yu%2E+2023%2E+Molgrapher%3A+graph%2Dbased+visual+recognition+of+chemical+structures%2E+In+Pro)
24. Siddharth M Narayanan, James D Braza, Ryan-Rhys Griffiths, Albert Bou, Geemi Wellawatte, Mayk Caldas Ramos, Ludovico Mitchener, Samuel G Rodriques, and Andrew D White. 2025. Training a scientific reasoning model for chemistry. arXiv preprint arXiv:2506.17238. [find](https://scholar.google.com/scholar?q=Siddharth+M+Narayanan%2C+James+D+Braza%2C+Ryan%2DRhys+Griffiths%2C+Albert+Bou%2C+Geemi+Wellawatte%2C+Mayk+Caldas+Ramos%2C+Ludovico+Mitchener%2C+Samuel+G+Rodriques%2C+and+Andrew+D+White%2E+2025%2E+Training+a+scientific+reas)
25. OpenAI. 2024. Gpt-4o. Accessed: 2024-05-13. [find](https://scholar.google.com/scholar?q=OpenAI%2E+2024%2E+Gpt%2D4o%2E+Accessed%3A+2024%2D05%2D13%2E)
26. Yujie Qian, Jiang Guo, Zhengkai Tu, Zhening Li, Connor W Coley, and Regina Barzilay. 2023. Molscribe: robust molecular structure recognition with image-to-graph generation. Journal of Chemical Information and Modeling, 63(7):1925–1934. [find](https://scholar.google.com/scholar?q=Yujie+Qian%2C+Jiang+Guo%2C+Zhengkai+Tu%2C+Zhening+Li%2C+Connor+W+Coley%2C+and+Regina+Barzilay%2E+2023%2E+Molscribe%3A+robust+molecular+structure+recognition+with+image%2Dto%2Dgraph+generation%2E+Journal+of+Chemical+Informa)
27. Kohulan Rajan, Achim Zielesny, and Christoph Steinbeck. 2021. Decimer 1.0: deep learning for chemical image recognition using transformers. Journal of Cheminformatics, 13:1–16. [find](https://scholar.google.com/scholar?q=Kohulan+Rajan%2C+Achim+Zielesny%2C+and+Christoph+Steinbeck%2E+2021%2E+Decimer+1%2E0%3A+deep+learning+for+chemical+image+recognition+using+transformers%2E+Journal+of+Cheminformatics%2C+13%3A1%E2%80%9316%2E)
28. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, YK Li, Yang Wu, and 1 others. 2024. [Deepseekmath: Pushing the limits of mathematical reasoning in open language models](https://md.research.elamur.ai/free/en/papers/shao-2024-deepseekmath-pushing-limits-mathematical-reasoning/). arXiv preprint arXiv:2402.03300.
29. Haozhan Shen, Peng Liu, Jingcheng Li, Chunxin Fang, Yibo Ma, Jiajia Liao, Qiaoli Shen, Zilun Zhang, Kangjia Zhao, Qianqian Zhang, and 1 others. 2025. [Vlm-r1: A stable and generalizable r1-style large vision-language model](https://md.research.elamur.ai/free/en/papers/shen-2025-vlm-r1-stable-generalizable-r1/). arXiv preprint arXiv:2504.07615.
30. Qian Tan, Dongzhan Zhou, Peng Xia, Wanhao Liu, Wanli Ouyang, Lei Bai, Yuqiang Li, and Tianfan Fu. 2025. Chemmllm: Chemical multimodal large language model. arXiv preprint arXiv:2505.16326. — [arXiv:2505.16326](https://arxiv.org/abs/2505.16326) [find](https://scholar.google.com/scholar?q=Qian+Tan%2C+Dongzhan+Zhou%2C+Peng+Xia%2C+Wanhao+Liu%2C+Wanli+Ouyang%2C+Lei+Bai%2C+Yuqiang+Li%2C+and+Tianfan+Fu%2E+2025%2E+Chemmllm%3A+Chemical+multimodal+large+language+model%2E+arXiv+preprint+arXiv%3A2505%2E16326%2E+%E2%80%94+%5BarXiv%3A25)
31. Core Team, Zihao Yue, Zhenru Lin, Yifan Song, Weikun Wang, Shuhuai Ren, Shuhao Gu, Shicheng Li, Peidian Li, Liang Zhao, Lei Li, Kainan Bao, Hao Tian, Hailin Zhang, Gang Wang, Dawei Zhu, Cici, Chenhong He, Bowen Ye, and 55 others. 2025. [Mimo-vl technical report](https://md.research.elamur.ai/free/en/papers/team-2025-llm-mimo-vl-technical-report/). Preprint, arXiv:2506.03569.
32. Yijun Tian, Shaoyu Chen, Zhichao Xu, Yawei Wang, Jinhe Bi, Peng Han, and Wei Wang. 2025. Reinforcement mid-training. arXiv preprint arXiv:2509.24375. — [arXiv:2509.24375](https://arxiv.org/abs/2509.24375) [find](https://scholar.google.com/scholar?q=Yijun+Tian%2C+Shaoyu+Chen%2C+Zhichao+Xu%2C+Yawei+Wang%2C+Jinhe+Bi%2C+Peng+Han%2C+and+Wei+Wang%2E+2025%2E+Reinforcement+mid%2Dtraining%2E+arXiv+preprint+arXiv%3A2509%2E24375%2E+%E2%80%94+%5BarXiv%3A2509%2E24375%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2509%2E24)
33. Haozhe Wang, Haoran Que, Qixin Xu, Minghao Liu, Wangchunshu Zhou, Jiazhan Feng, Wanjun Zhong, Wei Ye, Tong Yang, Wenhao Huang, and 1 others. 2025a. Reverse-engineered reasoning for open-ended generation. arXiv preprint arXiv:2509.06160. — [arXiv:2509.06160](https://arxiv.org/abs/2509.06160) [find](https://scholar.google.com/scholar?q=Haozhe+Wang%2C+Haoran+Que%2C+Qixin+Xu%2C+Minghao+Liu%2C+Wangchunshu+Zhou%2C+Jiazhan+Feng%2C+Wanjun+Zhong%2C+Wei+Ye%2C+Tong+Yang%2C+Wenhao+Huang%2C+and+1+others%2E+2025a%2E+Reverse%2Dengineered+reasoning+for+open%2Dended+generati)
34. Weida Wang, Benteng Chen, Di Zhang, Wanhao Liu, Shuchen Pu, Ben Gao, Jin Zeng, Xiaoyong Wei, Tianshu Yu, Shuzhou Sun, and 1 others. 2025b. Chem-r: Learning to reason as a chemist. arXiv preprint arXiv:2510.16880. — [arXiv:2510.16880](https://arxiv.org/abs/2510.16880) [find](https://scholar.google.com/scholar?q=Weida+Wang%2C+Benteng+Chen%2C+Di+Zhang%2C+Wanhao+Liu%2C+Shuchen+Pu%2C+Ben+Gao%2C+Jin+Zeng%2C+Xiaoyong+Wei%2C+Tianshu+Yu%2C+Shuzhou+Sun%2C+and+1+others%2E+2025b%2E+Chem%2Dr%3A+Learning+to+reason+as+a+chemist%2E+arXiv+preprint+arXiv)
35. Weiyun Wang, Zhangwei Gao, Lixin Gu, Hengjun Pu, Long Cui, Xingguang Wei, Zhaoyang Liu, Linglin Jing, Shenglong Ye, Jie Shao, and 1 others. 2025c. [Internvl3](https://md.research.elamur.ai/free/en/papers/wang-2025-internvl3-5-advancing-open/). 5: Advancing open-source multimodal models in versatility, reasoning, and efficiency. arXiv preprint arXiv:2508.18265.
36. Yizhou Wang, Chen Tang, Han Deng, Jiabei Xiao, Jiaqi Liu, Jianyu Wu, Jun Yao, Pengze Li, Encheng Su, Lintao Wang, and 1 others. 2025d. Scireasoner: Laying the scientific reasoning ground across disciplines. arXiv preprint arXiv:2509.21320. — [arXiv:2509.21320](https://arxiv.org/abs/2509.21320) [find](https://scholar.google.com/scholar?q=Yizhou+Wang%2C+Chen+Tang%2C+Han+Deng%2C+Jiabei+Xiao%2C+Jiaqi+Liu%2C+Jianyu+Wu%2C+Jun+Yao%2C+Pengze+Li%2C+Encheng+Su%2C+Lintao+Wang%2C+and+1+others%2E+2025d%2E+Scireasoner%3A+Laying+the+scientific+reasoning+ground+across+discip)
37. Daniel S Wigh, Joe Arrowsmith, Alexander Pomberger, Kobi C Felton, and Alexei A Lapkin. 2024. Orderly: data sets and benchmarks for chemical reaction data. Journal of Chemical Information and Modeling, 64(9):3790–3798. [find](https://scholar.google.com/scholar?q=Daniel+S+Wigh%2C+Joe+Arrowsmith%2C+Alexander+Pomberger%2C+Kobi+C+Felton%2C+and+Alexei+A+Lapkin%2E+2024%2E+Orderly%3A+data+sets+and+benchmarks+for+chemical+reaction+data%2E+Journal+of+Chemical+Information+and+Modeling)
38. Zichen Zhu, Danyang Zhang, and 1 others. 2025d. Chemdfm-r: An chemical reasoner llm enhanced with atomized chemical knowledge. arXiv preprint arXiv:2507.21990. — [arXiv:2507.21990](https://arxiv.org/abs/2507.21990) [find](https://scholar.google.com/scholar?q=Zichen+Zhu%2C+Danyang+Zhang%2C+and+1+others%2E+2025d%2E+Chemdfm%2Dr%3A+An+chemical+reasoner+llm+enhanced+with+atomized+chemical+knowledge%2E+arXiv+preprint+arXiv%3A2507%2E21990%2E+%E2%80%94+%5BarXiv%3A2507%2E21990%5D%28https%3A%2F%2Farxiv%2Eorg%2Fa)
39. Zihan Zhao, Da Ma, Lu Chen, Liangtai Sun, Zihao Li, Hongshen Xu, Zichen Zhu, Su Zhu, Shuai Fan, Guodong Shen, and 1 others. 2024b. Chemdfm: Dialogue foundation model for chemistry. arXiv e-prints, pages arXiv–2401. [find](https://scholar.google.com/scholar?q=Zihan+Zhao%2C+Da+Ma%2C+Lu+Chen%2C+Liangtai+Sun%2C+Zihao+Li%2C+Hongshen+Xu%2C+Zichen+Zhu%2C+Su+Zhu%2C+Shuai+Fan%2C+Guodong+Shen%2C+and+1+others%2E+2024b%2E+Chemdfm%3A+Dialogue+foundation+model+for+chemistry%2E+arXiv+e%2Dprints%2C+pag)
40. Yi Yang, Xiaoxuan He, Hongkun Pan, Xiyan Jiang, Yan Deng, Xingtao Yang, Haoyu Lu, Dacheng Yin, Fengyun Rao, Minfeng Zhu, and 1 others. 2025. R1-onevision: Advancing generalized multimodal reasoning through cross-modal formalization. arXiv preprint arXiv:2503.10615. [find](https://scholar.google.com/scholar?q=Yi+Yang%2C+Xiaoxuan+He%2C+Hongkun+Pan%2C+Xiyan+Jiang%2C+Yan+Deng%2C+Xingtao+Yang%2C+Haoyu+Lu%2C+Dacheng+Yin%2C+Fengyun+Rao%2C+Minfeng+Zhu%2C+and+1+others%2E+2025%2E+R1%2Donevision%3A+Advancing+generalized+multimodal+reasoning+th)
41. Qiying Yu, Zheng Zhang, Ruofei Zhu, Yufeng Yuan, Xiaochen Zuo, Yu Yue, Weinan Dai, Tiantian Fan, Gaohong Liu, Lingjun Liu, and 1 others. 2025. [Dapo: An open-source llm reinforcement learning system at scale](https://md.research.elamur.ai/free/en/papers/yu-2025-dapo-open-source-llm-reinforcement/). arXiv preprint arXiv:2503.14476.
42. Di Zhang, Wei Liu, Qian Tan, Jingdan Chen, Hang Yan, Yuliang Yan, Jiatong Li, Weiran Huang, Xiangyu Yue, Wanli Ouyang, and 1 others. 2024. Chemllm: A chemical large language model. arXiv preprint arXiv:2402.06852. [find](https://scholar.google.com/scholar?q=Di+Zhang%2C+Wei+Liu%2C+Qian+Tan%2C+Jingdan+Chen%2C+Hang+Yan%2C+Yuliang+Yan%2C+Jiatong+Li%2C+Weiran+Huang%2C+Xiangyu+Yue%2C+Wanli+Ouyang%2C+and+1+others%2E+2024%2E+Chemllm%3A+A+chemical+large+language+model%2E+arXiv+preprint+arXi)
43. Qiang Zhang, Keyan Ding, Tianwen Lv, Xinda Wang, Qingyu Yin, Yiwen Zhang, Jing Yu, Yuhao Wang, Xiaotong Li, Zhuoyi Xiang, and 1 others. 2025. Scientific large language models: A survey on biological & chemical domains. ACM Computing Surveys, 57(6):1–38. [find](https://scholar.google.com/scholar?q=Qiang+Zhang%2C+Keyan+Ding%2C+Tianwen+Lv%2C+Xinda+Wang%2C+Qingyu+Yin%2C+Yiwen+Zhang%2C+Jing+Yu%2C+Yuhao+Wang%2C+Xiaotong+Li%2C+Zhuoyi+Xiang%2C+and+1+others%2E+2025%2E+Scientific+large+language+models%3A+A+survey+on+biological+%26)
44. Guojiang Zhao, Sihang Li, Zixiang Lu, Zheng Cheng, Haitao Lin, Lirong Wu, Hanchen Xia, Hengxing Cai, Wentao Guo, Hongshuai Wang, and 1 others. 2025a. Molreasoner: Toward effective and interpretable reasoning for molecular llms. arXiv preprint arXiv:2508.02066. — [arXiv:2508.02066](https://arxiv.org/abs/2508.02066) [find](https://scholar.google.com/scholar?q=Guojiang+Zhao%2C+Sihang+Li%2C+Zixiang+Lu%2C+Zheng+Cheng%2C+Haitao+Lin%2C+Lirong+Wu%2C+Hanchen+Xia%2C+Hengxing+Cai%2C+Wentao+Guo%2C+Hongshuai+Wang%2C+and+1+others%2E+2025a%2E+Molreasoner%3A+Toward+effective+and+interpretable+re)
45. Xuanle Zhao, Xianzhen Luo, Qi Shi, Chi Chen, Shuo Wang, Zhiyuan Liu, and Maosong Sun. 2025b. [Chartcoder: Advancing multimodal large language model for chart-to-code generation](https://md.research.elamur.ai/free/en/papers/zhao-2025-chartcoder-advancing-multimodal-large-language/). arXiv preprint arXiv:2501.06598.
46. Xuanle Zhao, Shuxin Zeng, Yinyuan Cai, Xiang Cheng, Duzhen Zhang, Xiuyi Chen, and Bo Xu. 2025c. Tinychemvl: Advancing chemical vision-language models via efficient visual token reduction and complex reaction tasks. arXiv preprint arXiv:2511.06283. — [arXiv:2511.06283](https://arxiv.org/abs/2511.06283) [find](https://scholar.google.com/scholar?q=Xuanle+Zhao%2C+Shuxin+Zeng%2C+Yinyuan+Cai%2C+Xiang+Cheng%2C+Duzhen+Zhang%2C+Xiuyi+Chen%2C+and+Bo+Xu%2E+2025c%2E+Tinychemvl%3A+Advancing+chemical+vision%2Dlanguage+models+via+efficient+visual+token+reduction+and+complex+r)
47. Zihan Zhao, Bo Chen, Jingpiao Li, Lu Chen, Liyang Wen, Pengyu Wang, Zichen Zhu, Danyang Zhang, Yansi Li, Zhongyang Dai, and 1 others. 2024a. Chemdfm-x: towards large multimodal model for chemistry. Science China Information Sciences, 67(12):1–2. [find](https://scholar.google.com/scholar?q=Zihan+Zhao%2C+Bo+Chen%2C+Jingpiao+Li%2C+Lu+Chen%2C+Liyang+Wen%2C+Pengyu+Wang%2C+Zichen+Zhu%2C+Danyang+Zhang%2C+Yansi+Li%2C+Zhongyang+Dai%2C+and+1+others%2E+2024a%2E+Chemdfm%2Dx%3A+towards+large+multimodal+model+for+chemistry%2E+Sc)
48. Zihan Zhao, Bo Chen, Ziping Wan, Lu Chen, Xuanze Lin, Shiyang Yu, Situo Zhang, Da Ma, [find](https://scholar.google.com/scholar?q=Zihan+Zhao%2C+Bo+Chen%2C+Ziping+Wan%2C+Lu+Chen%2C+Xuanze+Lin%2C+Shiyang+Yu%2C+Situo+Zhang%2C+Da+Ma%2C)

*Updated 2 November 2026*