Elamur Research⌕
☰
home / papers / 2024 / Large Language Models as Biomedical Hypothesis Generators: A Comprehensive Evaluation

Large Language Models as Biomedical Hypothesis Generators: A Comprehensive Evaluation

Paper · 2024 · Biqing Qi, Kaiyan Zhang, Kai Tian et al.

Summary

The paper asks whether large language models can propose genuinely new biomedical hypotheses, rather than repeating knowledge absorbed during pre-training. Testing this is hard because the training corpora of most LLMs cannot be traced. The authors address it with a time-based split. They start from roughly 10,000 PubMed papers and use ChatGPT and GPT-4 to extract background-hypothesis pairs, following a Self-Instruct-style pipeline. Papers published before January 2023 form the training and "seen" portions, while papers from August 2023 form an "unseen" test set. Only models trained on data predating March 2023 are evaluated, in zero-shot, few-shot and fine-tuned modes. Outputs are scored with four newly proposed quality metrics, judged by both GPT-4 and humans. The study also tests a multi-agent framework and PubMed search tools. The main finding is that LLMs can produce novel hypotheses that later literature supports. Adding uncertainty through agent collaboration improves results, whereas extra retrieved knowledge does not reliably help. This provides a contamination-aware benchmark for LLM-assisted discovery.

Key points

  • Dataset: the seen set has 2,700 pairs, split into 2,500 for training and 200 for validation/seen testing. The unseen test set has 200 pairs from August 2023 papers. Data and code are available at TsinghuaC3I/LLM4BioHypoGen.
  • Models compared:
    • ChatGPT;
    • Llama-based general instruction models ranked highly on the Alpaca Eval Leaderboard;
    • the medical models PMC-LLaMA and MedAlpaca.
    • Few-shot prompts use 5 examples, chosen either randomly or by similarity retrieval.
  • Evaluation: GPT-4 ratings correlate with human ratings, which suggests LLMs can serve as hypothesis evaluators. In one showcased case, a fine-tuned 65B LLaMA predicted a metformin and esophageal cancer link that a later paper confirmed.
  • Multi-agent setup: the agents take the roles of Analyst, Scientist, Critic and Engineer, with the Engineer querying PubMed filtered to pre-2023 results. Agent collaboration clearly boosts zero-shot performance.
  • Tool use: tools alone add little, and ReAct performs slightly worse than OpenAI function calling. However, tools do further improve the multi-agent system.

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

🎓University of WashingtonSeattle · United States353 works · 200,043 citations🎓Tsinghua UniversityBeijing · China333 works · 87,019 citations🎓Harbin Institute of TechnologyHarbin · China158 works · 28,841 citations🔬Shanghai Artificial Intelligence LaboratoryShanghai · China117 works · 11,281 citations

Also on the byline: First Affiliated Hospital of Nanchang University.

 Fields, topics and concepts

📚 Shelves 8The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksBiochemistry, Genetics and Molecular Biology · 234 worksMolecular Biology · 231 worksMedicine · 241 worksHealth Informatics · 184 works

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

Cite

Cite as Biqing Qi, Kaiyan Zhang, Kai Tian, Haoxiang Li, Zhang-Ren Chen, Sihang Zeng, Ermo Hua, Hu Jinfang and Bowen Zhou. 2024. Large Language Models as Biomedical Hypothesis Generators: A Comprehensive Evaluation. arXiv:2407.08940.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2407.08940 · arXiv 2407.08940

@misc{qi2024large,
  title = {{Large Language Models as Biomedical Hypothesis Generators: A Comprehensive Evaluation}},
  author = {Biqing Qi, Kaiyan Zhang, Kai Tian, Haoxiang Li, Zhang-Ren Chen, Sihang Zeng, Ermo Hua, Hu Jinfang and Bowen Zhou},
  year = {2024},
  doi = {10.48550/arxiv.2407.08940},
  eprint = {2407.08940},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2407.08940}
}

References

58 references, 32 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Danial Alihosseini, Ehsan Montahaei, and Mahdieh Soleymani Baghshah. Jointly measuring diversity and quality in text generation models. In Proceedings of the Workshop on Methods for Optimizing and Evaluating Neural Language Generation, pp. 90–98, Minneapolis, Minnesota, jun 2019. Association for Com… find
  2. Daniil A Boiko, Robert MacKnight, and Gabe Gomes. Emergent autonomous scientific research capabilities of large language models. arXiv preprint arXiv:2304.05332, 2023. find
  3. Andres M. Bran, Sam Cox, Andrew D. White, and Philippe Schwaller. ChemCrow: Augmenting large-language models with chemistry tools, June 2023. URL http://arxiv.org/ abs/2304.05376. arXiv:2304.05376. ↗
  4. Boxi Cao, Hongyu Lin, Xianpei Han, and Le Sun. The Life Cycle of Knowledge in Big Language Models: A Survey, March 2023. URL http://arxiv.org/abs/2303.07616. arXiv:2303.07616 [cs]. ↗
  5. Zhuo Chang, Jing Zhang, Yilun Liu, Huajian Gao, and Guang-Kui Xu. New Mechanical Markers for Tracking the Progression of Myocardial Infarction. Nano Letters, 23(16): 7350–7357, August 2023. ISSN 1530-6984. doi: 10.1021/acs.nanolett.3c01712. — [doi:10.1021/acs.nanolett.3c01712](https://doi.org/10.1021/acs.nanolett.3c01712) find
  6. Yulin Chen, Ning Ding, Hai-Tao Zheng, Zhiyuan Liu, Maosong Sun, and Bowen Zhou. Empowering Private Tutoring by Chaining Large Language Models, September 2023. URL http://arxiv.org/abs/2309.08112. arXiv:2309.08112 [cs] version: 1. — [arXiv:2309.08112](https://arxiv.org/abs/2309.08112) ↗
  7. Yao Fu, Hao Peng, Tushar Khot, and Mirella Lapata. Improving Language Model Negotiation with Self-Play and In-Context Learning from AI Feedback, May 2023. URL http://arxiv.org/abs/2305.10142. arXiv:2305.10142 [cs]. ↗
  8. Fabrizio Gilardi, Meysam Alizadeh, and Maël Kubli. ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks. Proceedings of the National Academy of Sciences, 120(30): e2305016120, July 2023. ISSN 0027-8424, 1091-6490. doi: 10.1073/pnas.2305016120. URL http://arxiv.org/abs/2303.15056. arXiv:2303.… ↗
  9. Ran Gong, Qiuyuan Huang, Xiaojian Ma, Hoi Vo, Zane Durante, Yusuke Noda, Zilong Zheng, Song-Chun Zhu, Demetri Terzopoulos, Li Fei-Fei, and Jianfeng Gao. MindAgent: Emergent Gaming Interaction, September 2023. URL http://arxiv.org/abs/2309.09971. arXiv:2309.09971 [cs] version: 1. ↗
  10. Tianyu Han, Lisa C Adams, Jens-Michalis Papaioannou, Paul Grundmann, Tom Oberhauser, Alexander Löser, Daniel Truhn, and Keno K Bressem. Medalpaca–an open-source collection of medical conversational ai models and training data. arXiv preprint arXiv:2304.08247, 2023.
  11. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. Measuring massive multitask language understanding. arXiv preprint arXiv:2009.03300, 2020.
  12. Sirui Hong, Xiawu Zheng, Jonathan Chen, Yuheng Cheng, Jinlin Wang, Ceyao Zhang, Zili Wang, Steven Ka Shing Yau, Zijuan Lin, Liyang Zhou, Chenyu Ran, Lingfeng Xiao, and Chenglin Wu. MetaGPT: Meta Programming for Multi-Agent Collaborative Framework, August 2023. URL http://arxiv.org/abs/2308.00352. ar… ↗
  13. Chenxu Hu, Jie Fu, Chenzhuang Du, Simian Luo, Junbo Zhao, and Hang Zhao. ChatDB: Augmenting LLMs with Databases as Their Symbolic Memory, June 2023. URL http://arxiv.org/abs/2306.03901. arXiv:2306.03901 [cs]. — [arXiv:2306.03901](https://arxiv.org/abs/2306.03901) ↗
  14. Moksh Jain, Tristan Deleu, Jason Hartford, Cheng-Hao Liu, Alex Hernandez-Garcia, and Yoshua Bengio. Gflownets for ai-driven scientific discovery. Digital Discovery, 2(3): 557–577, 2023. find
  15. Albert Q Jiang, Alexandre Sablayrolles, Antoine Roux, Arthur Mensch, Blanche Savary, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Emma Bou Hanna, Florian Bressand, et al. Mixtral of experts. arXiv preprint arXiv:2401.04088, 2024. find
  16. Jikun Kang, Romain Laroche, Xindi Yuan, Adam Trischler, Xue Liu, and Jie Fu. Think Before You Act: Decision Transformers with Internal Working Memory, May 2023. URL http://arxiv.org/abs/2305.16338. 0 citations (Semantic Scholar/arXiv) [2023-05-30] arXiv:2305.16338 [cs]. — [arXiv:2305.16338](https://arxiv.org/abs/2305.16338) ↗
  17. Lorenz Kuhn, Yarin Gal, and Sebastian Farquhar. Semantic uncertainty: Linguistic invariances for uncertainty estimation in natural language generation. In NeurIPS ML Safety Workshop. Guohao Li, Hasan Abed Al Kader Hammoud, Hani Itani, Dmitrii Khizbullin, and Bernard Ghanem. CAMEL: Communicative Agen… — CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society
  18. Chin-Yew Lin. ROUGE: A package for automatic evaluation of summaries. In Text Summarization Branches Out, pp. 74–81, Barcelona, Spain, July 2004. Association for Computational Linguistics. URL https://aclanthology.org/W04-1013. ↗
  19. Yang Liu, Dan Iter, Yichong Xu, Shuohang Wang, Ruochen Xu, and Chenguang Zhu. G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment, May 2023. URL http://arxiv.org/abs/2303.16634. arXiv:2303.16634 [cs]. ↗
  20. Philipp Maas, Frank Carey, Chris Wheeler, Edward Saatchi, Pete Billington, and Jessica Yaffa Shamash. SHOW-1 and Showrunner Agents in Multi-Agent Simulations. arXiv preprint, 2023. find
  21. Reiichiro Nakano, Jacob Hilton, Suchir Balaji, Jeff Wu, Long Ouyang, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, Xu Jiang, Karl Cobbe, Tyna Eloundou, Gretchen Krueger, Kevin Button, Matthew Knight, Benjamin Chess, and John Schulman. WebGPT: Browser-assisted qu… — WebGPT: Browser-assisted question-answering with human feedback
  22. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th annual meeting of the Association for Computational Linguistics, pp. 311–318, 2002.
  23. Aaron Parisi, Yao Zhao, and Noah Fiedel. TALM: Tool Augmented Language Models, May 2022. URL http://arxiv.org/abs/2205.12255. arXiv:2205.12255 [cs]. ↗
  24. Joon Sung Park, Joseph C. O’Brien, Carrie J. Cai, Meredith Ringel Morris, Percy Liang, and Michael S. Bernstein. Generative Agents: Interactive Simulacra of Human Behavior, August 2023. URL http://arxiv.org/abs/2304.03442. arXiv:2304.03442 [cs]. ↗
  25. Shishir G. Patil, Tianjun Zhang, Xin Wang, and Joseph E. Gonzalez. Gorilla: Large Language Model Connected with Massive APIs, May 2023. URL http://arxiv.org/abs/2305.15334. arXiv:2305.15334 [cs]. ↗
  26. Chen Qian, Xin Cong, Wei Liu, Cheng Yang, Weize Chen, Yusheng Su, Yufan Dang, Jiahao Li, Juyuan Xu, Dahai Li, Zhiyuan Liu, and Maosong Sun. Communicative Agents for Software Development, August 2023. URL http://arxiv.org/abs/2307.07924. arXiv:2307.07924 [cs]. ↗
  27. Vipula Rawte, Amit Sheth, and Amitava Das. A Survey of Hallucination in Large Foundation Models, September 2023. URL http://arxiv.org/abs/2309.05922. arXiv:2309.05922 [cs]. ↗
  28. Timo Schick, Jane Dwivedi-Yu, Roberto Dessı̀, Roberta Raileanu, Maria Lomeli, Luke Zettlemoyer, Nicola Cancedda, and Thomas Scialom. Toolformer: Language models can teach themselves to use tools. arXiv preprint arXiv:2302.04761, 2023a.
  29. Timo Schick, Jane Dwivedi-Yu, Roberto Dessı̀, Roberta Raileanu, Maria Lomeli, Luke Zettlemoyer, Nicola Cancedda, and Thomas Scialom. Toolformer: Language Models Can Teach Themselves to Use Tools, February 2023b. URL http://arxiv.org/abs/2302.04761. arXiv:2302.04761. ↗
  30. John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347, 2017. find
  31. Yongliang Shen, Kaitao Song, Xu Tan, Dongsheng Li, Weiming Lu, and Yueting Zhuang. HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face, May 2023. URL http://arxiv.org/abs/2303.17580. arXiv:2303.17580 [cs]. ↗
  32. Noah Shinn, Federico Cassano, Beck Labash, Ashwin Gopinath, Karthik Narasimhan, and Shunyu Yao. Reflexion: Language Agents with Verbal Reinforcement Learning, June 2023. URL http://arxiv.org/abs/2303.11366. arXiv:2303.11366 [cs]. ↗
  33. Yifan Song, Weimin Xiong, Dawei Zhu, Cheng Li, Ke Wang, Ye Tian, and Sujian Li. RestGPT: Connecting Large Language Models with Real-World Applications via RESTful APIs, June 2023. URL http://arxiv.org/abs/2306.06624. arXiv:2306.06624 [cs]. ↗
  34. Ross Taylor, Marcin Kardas, Guillem Cucurull, Thomas Scialom, Anthony Hartshorn, Elvis Saravia, Andrew Poulton, Viktor Kerkez, and Robert Stojnic. Galactica: A large language model for science. arXiv preprint arXiv:2211.09085, 2022.
  35. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, et al. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971, 2023a.
  36. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288, 2023b.
  37. Karthik Valmeekam, Matthew Marquez, Sarath Sreedharan, and Subbarao Kambhampati. On the Planning Abilities of Large Language Models – A Critical Investigation, May 2023. URL http://arxiv.org/abs/2305.15771. arXiv:2305.15771 [cs]. ↗
  38. Laurens Van der Maaten and Geoffrey Hinton. Visualizing data using t-sne. Journal of machine learning research, 9(11), 2008. find
  39. Hanchen Wang, Tianfan Fu, Yuanqi Du, Wenhao Gao, Kexin Huang, Ziming Liu, Payal Chandak, Shengchao Liu, Peter Van Katwyk, Andreea Deac, et al. Scientific discovery in the age of artificial intelligence. Nature, 620(7972):47–60, 2023a. find
  40. Qingyun Wang, Doug Downey, Heng Ji, and Tom Hope. Learning to generate novel scientific directions with contextualized literature-based discovery. arXiv preprint arXiv:2305.14259, 2023b. find
  41. Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A Smith, Daniel Khashabi, and Hannaneh Hajishirzi. Self-instruct: Aligning language model with self generated instructions. arXiv preprint arXiv:2212.10560, 2022. find
  42. Christopher JCH Watkins and Peter Dayan. Q-learning. Machine learning, 8:279–292, 1992. find
  43. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. Chain-of-thought prompting elicits reasoning in large language models. Advances in Neural Information Processing Systems, 35:24824–24837, 2022.
  44. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. Chain-of-Thought Prompting Elicits Reasoning in Large Language Models, January 2023a. URL http://arxiv.org/abs/2201.11903. arXiv:2201.11903 [cs]. ↗
  45. Tianwen Wei, Liang Zhao, Lichang Zhang, Bo Zhu, Lijie Wang, Haihua Yang, Biye Li, Cheng Cheng, Weiwei Lü, Rui Hu, et al. Skywork: A more open bilingual foundation model. arXiv preprint arXiv:2310.19341, 2023b.
  46. Chaoyi Wu, Weixiong Lin, Xiaoman Zhang, Ya Zhang, Yanfeng Wang, and Weidi Xie. Pmc-llama: Towards building open-source language models for medicine, 2023.
  47. Zhiheng Xi, Wenxiang Chen, Xin Guo, Wei He, Yiwen Ding, Boyang Hong, Ming Zhang, Junzhe Wang, Senjie Jin, Enyu Zhou, Rui Zheng, Xiaoran Fan, Xiao Wang, Limao Xiong, Yuhao Zhou, Weiran Wang, Changhao Jiang, Yicheng Zou, Xiangyang Liu, Zhangyue Yin, Shihan Dou, Rongxiang Weng, Wensen Cheng, Qi Zhang,… — The Rise and Potential of Large Language Model Based Agents: A Survey
  48. Zhiheng Xi, Wenxiang Chen, Xin Guo, Wei He, Yiwen Ding, Boyang Hong, Ming Zhang, Junzhe Wang, Senjie Jin, Enyu Zhou, et al. The rise and potential of large language model based agents: A survey. arXiv preprint arXiv:2309.07864, 2023b.
  49. Miao Xiong, Zhiyuan Hu, Xinyang Lu, YIFEI LI, Jie Fu, Junxian He, and Bryan Hooi. Can llms express their uncertainty? an empirical evaluation of confidence elicitation in llms. In The Twelfth International Conference on Learning Representations. Hui Yang, Sifu Yue, and Yunzhong He. Auto-GPT for Onli… find
  50. Zonglin Yang, Xinya Du, Junxian Li, Jie Zheng, Soujanya Poria, and Erik Cambria. Large language models for automated open-domain scientific hypotheses discovery. arXiv preprint arXiv:2309.02726, 2023b. find
  51. Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L. Griffiths, Yuan Cao, and Karthik Narasimhan. Tree of Thoughts: Deliberate Problem Solving with Large Language Models, May 2023a. URL http://arxiv.org/abs/2305.10601. arXiv:2305.10601 [cs]. ↗
  52. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. ReAct: Synergizing Reasoning and Acting in Language Models, March 2023b. URL http://arxiv.org/abs/2210.03629. arXiv:2210.03629. ↗
  53. Kaiyan Zhang, Sihang Zeng, Ermo Hua, Ning Ding, Zhang-Ren Chen, Zhiyuan Ma, Haoxin Li, Ganqu Cui, Biqing Qi, Xuekai Zhu, et al. Ultramedical: Building specialized generalists in biomedicine. arXiv preprint arXiv:2406.03949, 2024. — [arXiv:2406.03949](https://arxiv.org/abs/2406.03949) find
  54. Yue Zhang, Yafu Li, Leyang Cui, Deng Cai, Lemao Liu, Tingchen Fu, Xinting Huang, Enbo Zhao, Yu Zhang, Yulong Chen, Longyue Wang, Anh Tuan Luu, Wei Bi, Freda Shi, and Shuming Shi. Siren’s Song in the AI Ocean: A Survey on Hallucination in Large Language Models, September 2023. URL — Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models http://arxiv.org/ab… ↗
  55. Ruiqi Zhong, Peter Zhang, Steve Li, Jinwoo Ahn, Dan Klein, and Jacob Steinhardt. Goal driven discovery of distributional differences via language descriptions. arXiv preprint arXiv:2302.14233, 2023.
  56. Kun Zhou, Yutao Zhu, Zhipeng Chen, Wentong Chen, Wayne Xin Zhao, Xu Chen, Yankai Lin, Ji-Rong Wen, and Jiawei Han. Don’t make your llm an evaluation benchmark cheater. arXiv preprint arXiv:2311.01964, 2023. find
  57. Xizhou Zhu, Yuntao Chen, Hao Tian, Chenxin Tao, Weijie Su, Chenyu Yang, Gao Huang, Bin Li, Lewei Lu, Xiaogang Wang, Yu Qiao, Zhaoxiang Zhang, and Jifeng Dai. Ghost in the Minecraft: Generally Capable Agents for Open-World Environments via Large Language Models with Text-based Knowledge and Memory, J… find
  58. Xuekai Zhu, Biqing Qi, Kaiyan Zhang, Xingwei Long, and Bowen Zhou. Pad: Program-aided distillation specializes large models in reasoning. arXiv preprint arXiv:2305.13888, 2023b. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/qi-2024-large-language-models-biomedical-hypothesis/