Elamur Research⌕
☰
home / papers / 2026 / AIDA-SEAT: Towards Reliable AI Doctor Assistant via State-Evaluation-Action Tree Enhanced LLMs in Online Hospital

AIDA-SEAT: Towards Reliable AI Doctor Assistant via State-Evaluation-Action Tree Enhanced LLMs in Online Hospital

Paper · 2026 · Lianxin Sun, Xiaoying Ying, Guangya Yu et al.

Summary

Clinical assistants built on LLMs, whether prompted directly or grounded through text or graph retrieval (RAG), mostly receive isolated medical facts. They get no explicit decision pathway to follow during multi-step clinical reasoning. The authors address this with AIDA-SEAT, a doctor-assistant system for an online hospital that is guided by State-Evaluation-Action Trees (SEAT). Each tree has root, condition, transition and action nodes, and its edges carry natural-language labels. The trees start as raw clinical guidance trees produced with the tool of Li et al. (2023). An LLM then rewrites node content as structured JSON logic, and clinicians correct the result. When a patient query arrives, the system works in three steps:

  • it picks a matching tree using LLM fact checks and a severity-based priority rule;
  • it walks the tree by labelling each branch condition;
  • it chooses between a SEAT-guided prompt and a plain CoT prompt to draft a reply, which a physician reviews before sending.

On an in-house dataset from five departments, the method beats CoT and RAG baselines by 1.01%. In live deployment it cut physicians' response time. This shows that encoding explicit clinical logic can make medical assistants both faster and more trustworthy.

Key points

  • Test data: an in-house set of 843 cases, with 200 each from Cardiology, Liver Surgery, Gastroenterology and General Medicine, plus 43 from Reproductive Medicine. Each department has its own symptom, examination and treatment trees.
  • Baselines: CoT, MedRAG, i-MedRAG and MedCite. SEAT-guided generation is 1.01% better than the strongest of them.
  • Branch logic: branch labels are judged True, False or Not Sure. When several trees or branches match, treatment trees take precedence over examination trees, and examination trees over symptom trees.
  • Deployment scale: the system has run in production since May 21, 2025. By January 4, 2026, 85 physicians had used it, with over 4,700 calls and more than 5,900 processed images.
  • Online results:
    • Across 200 consultations, assisted replies were 24.16 s faster, a 26.85% efficiency gain.
    • Average end-to-end latency is 6.43 s.
    • 85.7% of surveyed doctors approved of its clinical logic.
    • The cardiology chief physician's monthly consultations rose from 234 to 303, about 29.48%.

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

🎓Shanghai Jiao Tong UniversityShanghai · China149 works · 16,161 citations🎓East China University of Science and TechnologyShanghai · China10 works · 87 citations

Also on the byline: Renji Hospital.

 Fields, topics and concepts

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksMedicine · 241 worksHealth Informatics · 184 worksHealth Professions · 69 worksHealth Information Management · 50 works

🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 3,320 works

Cite

Cite as Lianxin Sun, Xiaoying Ying, Guangya Yu, Weiyan Zhang, Chenhao Guan, Hao He, Mingxi Shang, Jianhua Li, ChunMing Wang and Tong Ruan. 2026. AIDA-SEAT: Towards Reliable AI Doctor Assistant via State-Evaluation-Action Tree Enhanced LLMs in Online Hospital. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-industry.48

@inproceedings{sun2026aida,
  title = {{AIDA-SEAT: Towards Reliable AI Doctor Assistant via State-Evaluation-Action Tree Enhanced LLMs in Online Hospital}},
  author = {Lianxin Sun, Xiaoying Ying, Guangya Yu, Weiyan Zhang, Chenhao Guan, Hao He, Mingxi Shang, Jianhua Li, ChunMing Wang and Tong Ruan},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-industry.48},
  url = {https://doi.org/10.18653/v1/2026.acl-industry.48}
}

References

30 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, and 1 others. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774.
  2. Muhammad Arslan, Hussam Ghanem, Saba Munawar, and Christophe Cruz. 2024. A survey on rag with llms. Procedia computer science, 246:3781–3790. find
  3. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, and 1 others. 2025. Qwen2. 5-vl technical report. arXiv preprint arXiv:2502.13923. — Qwen2.5-VL Technical Report
  4. Junying Chen, Xidong Wang, Ke Ji, Anningzhe Gao, Feng Jiang, Shunian Chen, Hongbo Zhang, Song Dingjie, Wenya Xie, Chuyi Kong, Jianquan Li, Xiang Wan, Haizhou Li, and Benyou Wang. 2024. HuatuoGPT-II, one-stage training for medical adaption of LLMs. In First Conference on Language Modeling. find
  5. Chengfeng Dou, Ying Zhang, Zhi Jin, Wenpin Jiao, Haiyan Zhao, Yongqiang Zhao, and Zhengwei Tao. 2024. Integrating physician diagnostic logic into large language models: Preference learning from process feedback. In Findings of the Association for Computational Linguistics ACL 2024, pages 2453– 2473. find
  6. Darren Edge, Ha Trinh, Newman Cheng, Joshua Bradley, Alex Chao, Apurva Mody, Steven Truitt, Dasha Metropolitansky, Robert Osazuwa Ness, and Jonathan Larson. 2024. From local to global: A graph rag approach to query-focused summarization. arXiv preprint arXiv:2404.16130. find
  7. Zhihao Fan, Lai Wei, Jialong Tang, Wei Chen, Wang Siyuan, Zhongyu Wei, and Fei Huang. 2025. Ai hospital: Benchmarking large language models in a multi-agent medical interaction simulator. In Proceedings of the 31st International Conference on Computational Linguistics, pages 10183–10213.
  8. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, and 1 others. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783.
  9. 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
  10. Qiao Jin, Won Kim, Qingyu Chen, Donald C Comeau, Lana Yeganova, W John Wilbur, and Zhiyong Lu. 2023. Medcpt: Contrastive pre-trained transformers with large-scale pubmed search logs for zero-shot biomedical information retrieval. Bioinformatics, 39(11):btad651. find
  11. Binbin Li, Tianxin Meng, Xiaoming Shi, Jie Zhai, and Tong Ruan. 2023. Meddm: Llm-executable clinical guidance tree for clinical decision-making. arXiv preprint arXiv:2312.02441. — [arXiv:2312.02441](https://arxiv.org/abs/2312.02441) find
  12. Kangenbei Liao, CHENG ZHONG, Wei Chen, Qianlong Liu, zhongyu wei, Baolin Peng, and Xuanjing Huang. 2022. Task-oriented dialogue system for automatic disease diagnosis via hierarchical reinforcement learning. find
  13. Chengyuan Liu, Yangyang Kang, Shihang Wang, Lizhi Qing, Fubang Zhao, Chao Wu, Changlong Sun, Kun Kuang, and Fei Wu. 2024. More than catastrophic forgetting: Integrating general capabilities for domain-specific llms. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Proce… find
  14. Harsha Nori, Yin Tat Lee, Sheng Zhang, Dean Carignan, Richard Edgar, Nicolo Fusi, Nicholas King, Jonathan Larson, Yuanzhi Li, Weishung Liu, and 1 others. 2023. Can generalist foundation models outcompete special-purpose tuning? case study in medicine. arXiv preprint arXiv:2311.16452. find
  15. Xiaoming Shi, Zeming Liu, Li Du, Yuxuan Wang, Hongru Wang, Yuhang Guo, Tong Ruan, Jie Xu, Xiaofan Zhang, and Shaoting Zhang. 2024. Medical dialogue system: A survey of categories, methods, evaluation and challenges. Findings of the Association for Computational Linguistics ACL 2024, pages 2840–2861. find
  16. Karan Singhal, Shekoofeh Azizi, Tao Tu, S Sara Mahdavi, Jason Wei, Hyung Won Chung, Nathan Scales, Ajay Tanwani, Heather Cole-Lewis, Stephen Pfohl, and 1 others. 2023. Large language models encode clinical knowledge. Nature, 620(7972):172–180.
  17. Cunxiang Wang, Xiaoze Liu, Yuanhao Yue, Qipeng Guo, Xiangkun Hu, Xiangru Tang, Tianhang Zhang, Cheng Jiayang, Yunzhi Yao, Xuming Hu, and 1 others. 2025a. Survey on factuality in large language models. ACM Computing Surveys, 58(1):1–37. find
  18. Xiao Wang, Mengjue Tan, Qiao Jin, Guangzhi Xiong, Yu Hu, Aidong Zhang, Zhiyong Lu, and Minjia Zhang. 2025b. Medcite: Can language models generate verifiable text for medicine? arXiv preprint arXiv:2506.06605. find
  19. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems, 35:24824– 24837.
  20. Junde Wu, Jiayuan Zhu, Yunli Qi, Jingkun Chen, Min Xu, Filippo Menolascina, Yueming Jin, and Vicente Grau. 2025. Medical graph RAG: Evidence-based medical large language model via graph retrieval-augmented generation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Lin…
  21. Guangzhi Xiong, Qiao Jin, Zhiyong Lu, and Aidong Zhang. 2024a. Benchmarking retrieval-augmented generation for medicine. In Findings of the Association for Computational Linguistics ACL 2024, pages 6233–6251.
  22. Guangzhi Xiong, Qiao Jin, Xiao Wang, Minjia Zhang, Zhiyong Lu, and Aidong Zhang. 2024b. Improving retrieval-augmented generation in medicine with iterative follow-up questions. In Biocomputing 2025: Proceedings of the Pacific Symposium, pages 199– 214. World Scientific. find
  23. 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. arXiv preprint arXiv:2505.09388.
  24. An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, Junyang Lin, Kai Dang, and 22 others. 2024a. Qwen2.5 technical report. arXiv preprint arX… find
  25. Songhua Yang, Hanjie Zhao, Senbin Zhu, Guangyu Zhou, Hongfei Xu, Yuxiang Jia, and Hongying Zan. 2024b. Zhongjing: Enhancing the chinese medical capabilities of large language model through expert feedback and real-world multi-turn dialogue. In Proceedings of the AAAI conference on artificial intelli… find
  26. Jiayi Ye, Yanbo Wang, Yue Huang, Dongping Chen, Qihui Zhang, Nuno Moniz, Tian Gao, Werner Geyer, Chao Huang, Pin-Yu Chen, Nitesh V Chawla, and Xiangliang Zhang. 2025. Justice or prejudice? quantifying biases in LLM-as-a-judge. In The Thirteenth International Conference on Learning Representations.
  27. Xiechi Zhang, Zetian Ouyang, Linlin Wang, Gerard De Melo, Zhu Cao, Xiaoling Wang, Ya Zhang, Yanfeng Wang, and Liang He. 2025. AutoMedEval: Harnessing language models for automatic medical capability evaluation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistic… find
  28. Cheng Zhong, Kangenbei Liao, Wei Chen, Qianlong Liu, Baolin Peng, Xuanjing Huang, Jiajie Peng, and Zhongyu Wei. 2022. Hierarchical reinforcement learning for automatic disease diagnosis. Bioinformatics, 38(16):3995–4001. find
  29. Jiayuan Zhu, Jiazhen Pan, Yuyuan Liu, Fenglin Liu, and Junde Wu. 2025. Ask patients with patience: Enabling llms for human-centric medical dialogue with grounded reasoning. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 2846–2857.
  30. Wei Zhu, Wenfeng Li, Xiaoling Wang, Wendi Ji, Yuanbin Wu, Jin Chen, Liang Chen, and Buzhou Tang. 2022. Extracting decision trees from medical texts: an overview of the text2dt track in chip2022. In China Health Information Processing Conference, pages 89–102. Springer. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/sun-2026-aida-seat-reliable-ai-doctor/