Elamur Research⌕
☰
home / papers / 2026 / FARS: A Fully Automated Research System Deployed at Scale

FARS: A Fully Automated Research System Deployed at Scale

Paper · 2026 · Qiong Tang, Tianxiang Sun, Xiangkun Hu et al.

Summary

Automated research agents can already propose ideas, run experiments, and draft manuscripts. However, the evidence for this mostly comes from hand-picked examples or a few tasks framed by humans. This paper asks how such a system behaves when it runs continuously across many topics, and how its output should be judged. The authors build FARS, a pipeline of dedicated agents for Ideation, Planning, Experiment, and Writing. The agents coordinate through a shared workspace that serves as project memory and as an auditable archive of proposals, code, logs, results, and drafts. In its first public deployment on AI/ML research, FARS completed 166 papers across 67 fine-grained topics. The authors kept weak and negative outputs rather than presenting only the best ones. Volunteer reviewers wrote 282 structured reviews covering 140 papers. Quality varied widely. Some papers were judged review-worthy or even strong, but narrow experiments, methodological flaws, and integrity problems kept appearing. The work offers a large-scale, inspectable test of fully autonomous research, showing that generation scales while reliable contribution still lags.

Key points

  • The deployment produced 166 complete papers on 67 AI/ML topics. Some grew from nine seed topics given at launch, and the rest were emergent topics FARS found on its own. The agents ran on a cluster of 160 NVIDIA GPUs.
  • Reviews were collected from March 21 to April 12, 2026. They yielded 282 reviews for 140 papers, and the 26 papers without a review were excluded from the statistics.
  • Each review included an overall rating, sub-scores for soundness, presentation, and contribution, reviewer confidence, an integrity check, and disclosure of any LLM use.
  • The AI Integrity Audit had reviewers compare each manuscript with its code, logs, and raw outputs. It checked eight failure modes, such as hallucinated citations, fabricated results, spurious novelty, and experimental-design pathologies.
  • Three FARS papers (FA0328, FA0064, FA0029) were posted to arXiv after minimal human integrity review, with their content left unchanged.

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

🎓Stanford UniversityStanford · United States567 works · 335,190 citations🎓Carnegie Mellon UniversityPittsburgh · United States633 works · 195,218 citations🎓University of Illinois Urbana-ChampaignUrbana · United States228 works · 74,591 citations🎓National University of SingaporeSingapore · Singapore221 works · 39,953 citations🎓Fudan UniversityShanghai · China213 works · 22,907 citations🎓Hong Kong University of Science and TechnologyHong Kong · China183 works · 27,600 citations🏢TencentShenzhen · China174 works · 21,355 citations🎓Nanyang Technological UniversitySingapore · Singapore193 works · 27,739 citations🎓University of MichiganAnn Arbor · United States94 works · 18,548 citations🎓Chinese University of Hong KongHong Kong · China176 works · 17,536 citations🎓The Ohio State UniversityColumbus · United States96 works · 19,285 citations🏛️Chinese Academy of SciencesBeijing · China256 works · 25,136 citations🎓University of Hong KongPok Fu Lam · China127 works · 13,824 citations🔬Shanghai Artificial Intelligence LaboratoryShanghai · China164 works · 13,544 citations🎓McGill UniversityMontreal · Canada97 works · 11,755 citations🔬Institute of AutomationBeijing · China89 works · 14,908 citations🎓Sun Yat-sen UniversityGuangzhou · China80 works · 9,799 citations🏢ByteDanceBeijing · China54 works · 4,494 citations🎓East China Normal UniversityShanghai · China73 works · 2,653 citations🎓University of LeedsLeeds · United Kingdom43 works · 1,824 citations🎓Boston UniversityBoston · United States35 works · 4,458 citations🎓The Hong Kong University of Science and Technology (Guangzhou)Guangzhou · China73 works · 1,936 citations🎓University College DublinDublin · Ireland17 works · 1,888 citations🎓Washington University in St. LouisSt Louis · United States17 works · 1,159 citations🔬Helmholtz Center for Information SecuritySaarbrücken · Germany22 works · 1,197 citations🎓ShanghaiTech UniversityShanghai · China16 works · 1,991 citations🎓University of WarwickCoventry · United Kingdom9 works · 1,292 citations🏢ServiceNowSanta Clara · United States13 works · 980 citations

Also on the byline: Sun Yat-sen Memorial Hospital · Innovation Team.

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 19,503 worksArtificial Intelligence · 18,805 worksArts and Humanities · 17,959 worksLanguage and Linguistics · 17,789 worksHistory and Philosophy of Science · 83 works

Cite

Cite as Qiong Tang, Tianxiang Sun, Xiangkun Hu, Xiangyang Liu, Yiran Chen, Yunfan Shao, Bobo Li, Changze Lv, Cheng Xu, Chengsong Huang, Chunyang Li, Dizhan Xue, Hao Bai, Haodong Duan, Hengquan Guo, Hongyang He, Hongyi Chen, Hui Shen, Jiahao Yuan, Jiankai Sun, Jikang Cheng, Jinfeng Xu, Jingqi Tong, Jingye Chen, Jinxiu Liu, Jixuan Leng, Junchi Yu, Kaixun Jiang, Kun Xiang, Kunpeng Yao, Lang Feng, Liangqi Yuan, Longsen Gao, Meng Li, Qi Jia, Qiushi Sun, Shengyuan Ding, Shizhan Gong, Siru Zhong, Terry Jingchen Zhang, Tianle Gu, Tianyi Liang, Weijie Liu, Weikai Yang, Weizhi Fei, Xin Wang, Xinpeng Liu, Xuanwen Ding, Yihong Tang, Yuanli Wang, Yukun Jiang, Yuming Yang, Zhengbao He, Zhikai Chen, Zhikun Xu, Zhuang Li and Zihao Huang. 2026. FARS: A Fully Automated Research System Deployed at Scale. arXiv:2606.31651.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2606.31651 · arXiv 2606.31651

@misc{tang2026fars,
  title = {{FARS: A Fully Automated Research System Deployed at Scale}},
  author = {Qiong Tang, Tianxiang Sun, Xiangkun Hu, Xiangyang Liu, Yiran Chen, Yunfan Shao, Bobo Li, Changze Lv, Cheng Xu, Chengsong Huang, Chunyang Li, Dizhan Xue, Hao Bai, Haodong Duan, Hengquan Guo, Hongyang He, Hongyi Chen, Hui Shen, Jiahao Yuan, Jiankai Sun, Jikang Cheng, Jinfeng Xu, Jingqi Tong, Jingye Chen, Jinxiu Liu, Jixuan Leng, Junchi Yu, Kaixun Jiang, Kun Xiang, Kunpeng Yao, Lang Feng, Liangqi Yuan, Longsen Gao, Meng Li, Qi Jia, Qiushi Sun, Shengyuan Ding, Shizhan Gong, Siru Zhong, Terry Jingchen Zhang, Tianle Gu, Tianyi Liang, Weijie Liu, Weikai Yang, Weizhi Fei, Xin Wang, Xinpeng Liu, Xuanwen Ding, Yihong Tang, Yuanli Wang, Yukun Jiang, Yuming Yang, Zhengbao He, Zhikai Chen, Zhikun Xu, Zhuang Li and Zihao Huang},
  year = {2026},
  doi = {10.48550/arxiv.2606.31651},
  eprint = {2606.31651},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2606.31651}
}

References

34 references, 21 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Anthropic. Effective harnesses for long-running agents. https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents, 2025. Anthropic Engineering Blog. ↗
  2. Akari Asai, Jacqueline He, Rulin Shao, Weijia Shi, Amanpreet Singh, Joseph Chee Chang, Kyle Lo, Luca Soldaini, Sergey Feldman, Mike D’arcy, David Wadden, Matt Latzke, Minyang Tian, Pan Ji, Shengyan Liu, Hao Tong, Bohao Wu, Yanyu Xiong, Luke Zettlemoyer, Graham Neubig, Dan Weld, Doug Downey, Wen tau… — OpenScholar: Synthesizing Scientific Literature with Retrieval-augmented LMs
  3. Jinheon Baek, Sujay Kumar Jauhar, Silviu Cucerzan, and Sung Ju Hwang. Researchagent: Iterative research idea generation over scientific literature with large language models, 2025. URL https://arxiv.org/abs/2404.07738. ↗
  4. Alina Beygelzimer, Yann N. Dauphin, Percy Liang, and Jennifer Wortman Vaughan. Has the machine learning review process become more arbitrary as the field has grown? the neurips 2021 consistency experiment, 2023. URL https://arxiv.org/abs/2306.03262. ↗
  5. Hui Chen, Miao Xiong, Yujie Lu, Wei Han, Ailin Deng, Yufei He, Jiaying Wu, Yibo Li, Yue Liu, and Bryan Hooi. Mlr-bench: Evaluating ai agents on open-ended machine learning research, 2025. URL https://arxiv.org/abs/2505.19955. ↗
  6. Ali Essam Ghareeb, Benjamin Chang, Ludovico Mitchener, Angela Yiu, Caralyn J. Szostkiewicz, Dmytro Shved, Gavin J. Gyimesi, Jon M. Laurent, Samantha M. Wright, Muhammed T. Razzak, Andrew D. White, Silvia C. Finnemann, Michaela M. Hinks, and Samuel G. Rodriques. A multi-agent system for automating sc… — [doi:10.1038/s41586-026-10652-y](https://doi.org/10.1038/s41586-026-10652-y) find
  7. Google. Nano Banana Pro: Gemini 3 Pro Image model from Google DeepMind. https://blog.google/innovation-and-ai/products/nano-banana-pro/, 11 2025. Accessed: 2026-03-19. ↗
  8. ≥ 6 (%) Reviewer \# Papers Mean rating SAR SAR SAR SAR SAR SAR SAR 71 3 6 7 20 3 165 2.29 2.40 2.78 3.19 4.00 4.13 5.00 0.0 0.0 0.0 0.0 0.0 0.0 1.8 Human 3.23 11.4 Jiawei Gu, Xuhui Jiang, Zhichao Shi, Hexiang Tan, Xuehao Zhai, Chengjin Xu, Wei Li, Yinghan Shen, Shengjie Ma, Honghao Liu, Saizhuo Wang,… find
  9. Tarun Gupta and Danish Pruthi. All that glitters is not novel: Plagiarism in AI generated research. In Proceedings of ACL 2025, 2025.
  10. Qian Huang, Jian Vora, Percy Liang, and Jure Leskovec. Mlagentbench: Evaluating language agents on machine learning experimentation, 2024. URL https://arxiv.org/abs/2310.03302. ↗
  11. Intology. Zochi technical report. arXiv, 2025. find
  12. Peter Alexander Jansen, Oyvind Tafjord, Marissa Radensky, Pao Siangliulue, Tom Hope, Bhavana Dalvi, Bodhisattwa Prasad Majumder, Daniel S. Weld, and Peter Clark. CodeScientist: End-to-end semi-automated scientific discovery with code-based experimentation. In Findings of the Association for Computat…
  13. Amir Hossein Kargaran, Nafiseh Nikeghbal, Jing Yang, and Nedjma Ousidhoum. Insights from the iclr peer review and rebuttal process, 2025. URL https://arxiv.org/abs/2511.15462. — [arXiv:2511.15462](https://arxiv.org/abs/2511.15462) ↗
  14. Patrick Tser Jern Kon, Jiachen Liu, Xinyi Zhu, Qiuyi Ding, Jingjia Peng, Jiarong Xing, Yibo Huang, Yiming Qiu, Jayanth Srinivasa, Myungjin Lee, Mosharaf Chowdhury, Matei Zaharia, and Ang Chen. Exp-bench: Can ai conduct ai research experiments?, 2025. URL https://arxiv.org/abs/2505.24785. ↗
  15. Ruochen Li, Teerth Patel, Qingyun Wang, and Xinya Du. Mlr-copilot: Autonomous machine learning research based on large language models agents, 2025. URL https://arxiv.org/abs/2408.14033. ↗
  16. Yu Li, Chenyang Shao, Xinyang Liu, Ruotong Zhao, Peijie Liu, Hongyuan Su, Zhibin Chen, Qinglong Yang, Anjie Xu, Yi Fang, Qingbin Zeng, Tianxing Li, Jingbo Xu, Fengli Xu, Yong Li, and Tie-Yan Liu. AutoSOTA: An end-to-end automated research system for state-of-the-art AI model discovery. arXiv preprin…
  17. Weixin Liang, Yuhui Zhang, Hancheng Cao, Binglu Wang, Daisy Ding, Xinyu Yang, Kailas Vodrahalli, Siyu He, Daniel Smith, Yian Yin, Daniel McFarland, and James Zou. Can large language models provide useful feedback on research papers? a large-scale empirical analysis, 2023. URL https://arxiv.org/abs/2… ↗
  18. Jiaqi Liu, Shi Qiu, Mairui Li, Bingzhou Li, Haonian Ji, Siwei Han, Xinyu Ye, Peng Xia, Zihan Dong, Meng Chen, Congyu Zhang, Letian Zhang, Guiming Chen, Haoqin Tu, Xinyu Yang, Lu Feng, Xujiang Zhao, Haifeng Chen, Jiawei Zhou, Xiao Wang, Weitong Zhang, Hongtu Zhu, Yun Li, Jieru Mei, Hongliang Fei, Jia… — [arXiv:2605.20025](https://arxiv.org/abs/2605.20025) find
  19. Zijun Liu, Kaiming Liu, Yiqi Zhu, Xuanyu Lei, Zonghan Yang, Zhenhe Zhang, Peng Li, and Yang Liu. Aigs: Generating science from ai-powered automated falsification, 2024. URL https://arxiv.org/abs/2411.11910. — [arXiv:2411.11910](https://arxiv.org/abs/2411.11910) ↗
  20. Chris Lu, Cong Lu, Robert Tjarko Lange, Jakob Foerster, Jeff Clune, and David Ha. The ai scientist: Towards fully automated open-ended scientific discovery, 2024. URL https://arxiv.org/abs/2408.06292. ↗
  21. Rui Meng, Bhavana Dalvi Mishra, Jiefeng Chen, Chun-Liang Li, Palash Goyal, Mihir Parmar, Yiwen Song, Yale Song, Rajarishi Sinha, Parthasarathy Ranganathan, Burak Gokturk, Jinsung Yoon, and Tomas Pfister. Scientistone: Towards human-level autonomous research via chain-of-evidence, 2026. URL https://a… find
  22. David B. Resnik, Mohammad Hosseini, and Rico Hauswald. Autonomous artificial intelligence, scientific research, and human values. AI and Ethics, 2026. doi: 10.1007/s43681-025-00908-0. — [doi:10.1007/s43681-025-00908-0](https://doi.org/10.1007/s43681-025-00908-0) find
  23. Samuel Schmidgall, Yusheng Su, Ze Wang, Ximeng Sun, Jialian Wu, Xiaodong Yu, Jiang Liu, Michael Moor, Zicheng Liu, and Emad Barsoum. Agent laboratory: Using llm agents as research assistants, 2025. URL https://arxiv.org/abs/2501.04227. ↗
  24. Chenglei Si, Diyi Yang, and Tatsunori Hashimoto. Can llms generate novel research ideas? a large-scale human study with 100+ nlp researchers, 2024. URL https://arxiv.org/abs/2409.04109. ↗
  25. Chenglei Si, Tatsunori Hashimoto, and Diyi Yang. The ideation-execution gap: Execution outcomes of LLM-generated versus human research ideas. arXiv preprint arXiv:2506.20803, 2025.
  26. Giulio Starace, Oliver Jaffe, Dane Sherburn, James Aung, Jun Shern Chan, Leon Maksin, Rachel Dias, Evan Mays, Benjamin Kinsella, Wyatt Thompson, Johannes Heidecke, Amelia Glaese, and Tejal Patwardhan. PaperBench: Evaluating AI’s ability to replicate AI research. In Proceedings of the 42nd Internatio… — PaperBench: Evaluating AI's Ability to Replicate AI Research
  27. Jiabin Tang, Lianghao Xia, Zhonghang Li, and Chao Huang. Ai-researcher: Autonomous scientific innovation, 2025. URL https://arxiv.org/abs/2505.18705. ↗
  28. Yixuan Weng, Minjun Zhu, Guangsheng Bao, Hongbo Zhang, Jindong Wang, Yue Zhang, and Linyi Yang. Cycleresearcher: Improving automated research via automated review. In The Thirteenth International Conference on Learning Representations, 2025a. URL https://openreview.net/forum?id=bjcsVLoHYs. ↗
  29. Yixuan Weng, Minjun Zhu, Qiujie Xie, Qiyao Sun, Zhen Lin, Sifan Liu, and Yue Zhang. Deepscientist: Advancing frontier-pushing scientific findings progressively, 2025b. URL https://arxiv.org/abs/2509.26603. ↗
  30. Yutaro Yamada, Robert Tjarko Lange, Cong Lu, Shengran Hu, Chris Lu, Jakob Foerster, Jeff Clune, and David Ha. The ai scientist-v2: Workshop-level automated scientific discovery via agentic tree search, 2025. URL https://arxiv.org/abs/2504.08066. ↗
  31. Ruofeng Yang, Yongcan Li, and Shuai Li. Aris: Autonomous research via adversarial multi-agent collaboration, 2026. URL https://arxiv.org/abs/2605.03042. — [arXiv:2605.03042](https://arxiv.org/abs/2605.03042) ↗
  32. Jianxiang Yu, Zichen Ding, Jiaqi Tan, Kangyang Luo, Zhenmin Weng, Chenghua Gong, Long Zeng, Renjing Cui, Chengcheng Han, Qiushi Sun, Zhiyong Wu, Yunshi Lan, and Xiang Li. Automated peer reviewing in paper sea: Standardization, evaluation, and analysis, 2024. URL https://arxiv.org/abs/2407.12857. ↗
  33. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric P. Xing, Hao Zhang, Joseph E. Gonzalez, and Ion Stoica. Judging llm-as-a-judge with mt-bench and chatbot arena, 2023. URL https://arxiv.org/abs/2306.05685. ↗
  34. Kunlun Zhu, Jiaxun Zhang, Ziheng Qi, et al. SafeScientist: Toward risk-aware scientific discoveries by LLM agents. arXiv preprint arXiv:2505.23559, 2025.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/tang-2026-fars-fully-automated-research/