---
title: "Workflow Closure Is Not Scientific Closure in Auto-Research"
description: "Because these stem from design choices rather than limits of autonomy, the authors suggest remedies… 122 references, cited by 1 paper here."
authors: ["Shuai Wang", "Xinyuan Tian", "Pangpang Liu", "Yize Zhao"]
year: 2026
venue: "arXiv:2605.26200"
source: "https://arxiv.org/abs/2605.26200"
doi: "10.48550/arxiv.2605.26200"
arxiv: "2605.26200"
licence: "arXiv non-exclusive distribution license"
licence_url: "http://arxiv.org/licenses/nonexclusive-distrib/1.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/wang-2026-workflow-closure-scientific-closure/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/wang-2026-workflow-closure-scientific-closure/)

[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/20/) / Workflow Closure Is Not Scientific Closure in Auto-Research Systems

# Workflow Closure Is Not Scientific Closure in Auto-Research Systems

**Authors:** [Shuai Wang](https://md.research.elamur.ai/free/en/authors/shuai-wang/), Xinyuan Tian, Pangpang Liu and Yize Zhao · **Year:** 2026 · **Venue:** arXiv:2605.26200 · **Licence:** [arXiv non-exclusive distribution license](http://arxiv.org/licenses/nonexclusive-distrib/1.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2605.26200) · [PDF](https://arxiv.org/pdf/2605.26200) · [DOI 10.48550/arxiv.2605.26200](https://doi.org/10.48550/arxiv.2605.26200) · [arXiv 2605.26200](https://arxiv.org/abs/2605.26200) · [OpenAlex](https://openalex.org/W7162579537)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · [1](https://md.research.elamur.ai/free/en/papers/wang-2026-workflow-closure-scientific-closure/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary

This position paper asks what it actually means when an automated research system "closes the loop." The authors separate two notions. **Workflow closure** is a system's ability to run ideation, experiments, writing and evaluation without human input, judging itself with its own evaluators. **Scientific closure** requires outputs to face plural objectives, validators independent of the producer, and community uptake. They argue the first is a genuine engineering result but gives no scientific standing.

Drawing on a survey of over 100 papers and repositories and a structured audit of 21 documented systems, they identify three linked failures:

- **Objective collapse:** a single proxy metric drives the loop, a Goodhart-type problem.
- **Validation collapse:** checking stays within the same model family or benchmark regime.
- **Acceptance collapse:** loops end in scores or paper-like artifacts rather than open critique and reuse.

Because these stem from design choices rather than limits of autonomy, the authors suggest remedies and advocate autonomous execution under external, non-autonomous epistemic control.

## Key points

- The evidence base is a survey of more than 100 recent works and repositories, plus an audit of 21 systems coded on three dimensions (L1 objective, L2 validation, L3 acceptance).
- Each dimension is labeled using a three-way scheme:
  - **Strong:** the loop is governed only by internal signals.
  - **Weak:** external involvement is simulated from inside the system.
  - **Mitigated:** a truly external check binds the loop.
- No audited system earned a Mitigated label on any dimension. For example, Karpathy autoresearch, ADAS, AIDE and AI-Scientist-v2 are Strong on all three.
- SakanaAI/AI-Scientist, whose papers passed workshop peer review per the Nature publication, is coded S/W/W. AI-Researcher is the only system coded W/W/W.
- CycleResearcher uses averaged reviewer scores as a SimPO reward, produced by a same-family reward model trained on ICLR-2024 review data. This is an illustration of validation collapse.

*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

- [Yale University](https://md.research.elamur.ai/free/en/universities/yale-university/) — University, New Haven, United States · 112 works · 14,009 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 19,503 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 18,805 works · [🤝 Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · 2,009 works · [🤝 Safety Research](https://md.research.elamur.ai/free/en/shelves/social-sciences/safety-research/) · 665 works · [🎯 Decision Sciences](https://md.research.elamur.ai/free/en/shelves/decision-sciences/) · 306 works · [🎯 Information Systems and Management](https://md.research.elamur.ai/free/en/shelves/decision-sciences/information-systems-and-management/) · 96 works

## Cite

Cite as: Shuai Wang, Xinyuan Tian, Pangpang Liu and Yize Zhao. 2026. Workflow Closure Is Not Scientific Closure in Auto-Research Systems. arXiv:2605.26200.

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

```bibtex
@misc{wang2026workflow,
  title = {{Workflow Closure Is Not Scientific Closure in Auto-Research Systems}},
  author = {Shuai Wang, Xinyuan Tian, Pangpang Liu and Yize Zhao},
  year = {2026},
  doi = {10.48550/arxiv.2605.26200},
  eprint = {2605.26200},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2605.26200}
}
```

- [← previous Word Alignment-Based Evaluation of Uniform Meaning Representations](https://md.research.elamur.ai/free/en/papers/zeman-2026-word-alignment-based-evaluation-uniform/)
- [next → X-NLP at SemEval-2026 Task 12: Prompting LLMs for Abductive Event Reasoning](https://md.research.elamur.ai/free/en/papers/mattie-2026-x-nlp-semeval-2026-task/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/20/)

## References

122 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. du-nlp-lab/MLR-Copilot, Aug. 2025. [find](https://scholar.google.com/scholar?q=du%2Dnlp%2Dlab%2FMLR%2DCopilot%2C+Aug%2E+2025%2E)
2. aiming-lab/AutoResearchClaw, Apr. 2026. [find](https://scholar.google.com/scholar?q=aiming%2Dlab%2FAutoResearchClaw%2C+Apr%2E+2026%2E)
3. davebcn87/pi-autoresearch, Apr. 2026. [find](https://scholar.google.com/scholar?q=davebcn87%2Fpi%2Dautoresearch%2C+Apr%2E+2026%2E)
4. drivelineresearch/autoresearch-claude-code, Apr. 2026. [find](https://scholar.google.com/scholar?q=drivelineresearch%2Fautoresearch%2Dclaude%2Dcode%2C+Apr%2E+2026%2E)
5. eimenhmdt/autoresearcher, Apr. 2026. [find](https://scholar.google.com/scholar?q=eimenhmdt%2Fautoresearcher%2C+Apr%2E+2026%2E)
6. Entrpi/autoresearch-everywhere, Apr. 2026. [find](https://scholar.google.com/scholar?q=Entrpi%2Fautoresearch%2Deverywhere%2C+Apr%2E+2026%2E)
7. greyhaven-ai/autocontext, Apr. 2026. [find](https://scholar.google.com/scholar?q=greyhaven%2Dai%2Fautocontext%2C+Apr%2E+2026%2E)
8. HKUDS/AI-Researcher, Apr. 2026. [find](https://scholar.google.com/scholar?q=HKUDS%2FAI%2DResearcher%2C+Apr%2E+2026%2E)
9. HKUDS/ClawTeam, Apr. 2026. [find](https://scholar.google.com/scholar?q=HKUDS%2FClawTeam%2C+Apr%2E+2026%2E)
10. hyperspaceai/agi, Apr. 2026. [find](https://scholar.google.com/scholar?q=hyperspaceai%2Fagi%2C+Apr%2E+2026%2E)
11. james-s-tayler/lazy-developer, Apr. 2026. [find](https://scholar.google.com/scholar?q=james%2Ds%2Dtayler%2Flazy%2Ddeveloper%2C+Apr%2E+2026%2E)
12. JinheonBaek/ResearchAgent, Mar. 2026. [find](https://scholar.google.com/scholar?q=JinheonBaek%2FResearchAgent%2C+Mar%2E+2026%2E)
13. jmilinovich/goal-md, Apr. 2026. [find](https://scholar.google.com/scholar?q=jmilinovich%2Fgoal%2Dmd%2C+Apr%2E+2026%2E)
14. leo-lilinxiao/codex-autoresearch, Apr. 2026. [find](https://scholar.google.com/scholar?q=leo%2Dlilinxiao%2Fcodex%2Dautoresearch%2C+Apr%2E+2026%2E)
15. LitLLM/LitLLM, Apr. 2026. [find](https://scholar.google.com/scholar?q=LitLLM%2FLitLLM%2C+Apr%2E+2026%2E)
16. MASWorks/ML-Agent, Mar. 2026. [find](https://scholar.google.com/scholar?q=MASWorks%2FML%2DAgent%2C+Mar%2E+2026%2E)
17. MaximeRobeyns/self_improving_coding_agent, Apr. 2026. [find](https://scholar.google.com/scholar?q=MaximeRobeyns%2Fself%5Fimproving%5Fcoding%5Fagent%2C+Apr%2E+2026%2E)
18. metauto-ai/HGM, Apr. 2026. [find](https://scholar.google.com/scholar?q=metauto%2Dai%2FHGM%2C+Apr%2E+2026%2E)
19. MrTsepa/autoevolve, Mar. 2026. [find](https://scholar.google.com/scholar?q=MrTsepa%2Fautoevolve%2C+Mar%2E+2026%2E)
20. mutable-state-inc/autoresearch-at-home, Apr. 2026. [find](https://scholar.google.com/scholar?q=mutable%2Dstate%2Dinc%2Fautoresearch%2Dat%2Dhome%2C+Apr%2E+2026%2E)
21. openags/OpenAGS, Apr. 2026. [find](https://scholar.google.com/scholar?q=openags%2FOpenAGS%2C+Apr%2E+2026%2E)
22. OpenRaiser/NanoResearch, Apr. 2026. [find](https://scholar.google.com/scholar?q=OpenRaiser%2FNanoResearch%2C+Apr%2E+2026%2E)
23. Orchestra-Research/AI-Research-SKILLs, Apr. 2026. [find](https://scholar.google.com/scholar?q=Orchestra%2DResearch%2FAI%2DResearch%2DSKILLs%2C+Apr%2E+2026%2E)
24. peterskoett/self-improving-agent, Apr. 2026. [find](https://scholar.google.com/scholar?q=peterskoett%2Fself%2Dimproving%2Dagent%2C+Apr%2E+2026%2E)
25. PouriaRouzrokh/LatteReview, Apr. 2026. [find](https://scholar.google.com/scholar?q=PouriaRouzrokh%2FLatteReview%2C+Apr%2E+2026%2E)
26. SakanaAI/AI-Scientist, Apr. 2026. [find](https://scholar.google.com/scholar?q=SakanaAI%2FAI%2DScientist%2C+Apr%2E+2026%2E)
27. SakanaAI/AI-Scientist-v2, Apr. 2026. [find](https://scholar.google.com/scholar?q=SakanaAI%2FAI%2DScientist%2Dv2%2C+Apr%2E+2026%2E)
28. SamuelSchmidgall/AgentLaboratory, Apr. 2026. [find](https://scholar.google.com/scholar?q=SamuelSchmidgall%2FAgentLaboratory%2C+Apr%2E+2026%2E)
29. ShengranHu/ADAS, Apr. 2026. [find](https://scholar.google.com/scholar?q=ShengranHu%2FADAS%2C+Apr%2E+2026%2E)
30. Sibyl-Research-Team/AutoResearch-SibylSystem, Apr. 2026. [find](https://scholar.google.com/scholar?q=Sibyl%2DResearch%2DTeam%2FAutoResearch%2DSibylSystem%2C+Apr%2E+2026%2E)
31. supratikpm/gemini-autoresearch, Apr. 2026. [find](https://scholar.google.com/scholar?q=supratikpm%2Fgemini%2Dautoresearch%2C+Apr%2E+2026%2E)
32. uditgoenka/autoresearch, Apr. 2026. [find](https://scholar.google.com/scholar?q=uditgoenka%2Fautoresearch%2C+Apr%2E+2026%2E)
33. wanshuiyin/Auto-claude-code-research-in-sleep, Apr. 2026. [find](https://scholar.google.com/scholar?q=wanshuiyin%2FAuto%2Dclaude%2Dcode%2Dresearch%2Din%2Dsleep%2C+Apr%2E+2026%2E)
34. WecoAI/aideml, Apr. 2026. [find](https://scholar.google.com/scholar?q=WecoAI%2Faideml%2C+Apr%2E+2026%2E)
35. Why AI cannot do good science without humans. Nature, 653(8115):650–650, May 2026. [find](https://scholar.google.com/scholar?q=Why+AI+cannot+do+good+science+without+humans%2E+Nature%2C+653%288115%29%3A650%E2%80%93650%2C+May+2026%2E)
36. zkarimi22/autoresearch-anything, Apr. 2026. [find](https://scholar.google.com/scholar?q=zkarimi22%2Fautoresearch%2Danything%2C+Apr%2E+2026%2E)
37. S. Alexander, B. Bradley, L. Gouskos, and C. Niu. Autonomous Discovery of Particle Physics Theories from Experimental Data, Mar. 2026. [find](https://scholar.google.com/scholar?q=S%2E+Alexander%2C+B%2E+Bradley%2C+L%2E+Gouskos%2C+and+C%2E+Niu%2E+Autonomous+Discovery+of+Particle+Physics+Theories+from+Experimental+Data%2C+Mar%2E+2026%2E)
38. S. Alzubi, N. Provenzano, J. Bingham, W. Chen, and T. Vu. Evoskill: Automated skill discovery for multi-agent systems. arXiv preprint arXiv:2603.02766, 2026. — [arXiv:2603.02766](https://arxiv.org/abs/2603.02766) [find](https://scholar.google.com/scholar?q=S%2E+Alzubi%2C+N%2E+Provenzano%2C+J%2E+Bingham%2C+W%2E+Chen%2C+and+T%2E+Vu%2E+Evoskill%3A+Automated+skill+discovery+for+multi%2Dagent+systems%2E+arXiv+preprint+arXiv%3A2603%2E02766%2C+2026%2E+%E2%80%94+%5BarXiv%3A2603%2E02766%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs)
39. E. Aygün, A. Belyaeva, G. Comanici, M. Coram, H. Cui, J. Garrison, R. Johnston, A. Kast, C. Y. McLean, P. Norgaard, Z. Shamsi, D. Smalling, J. Thompson, S. Venugopalan, B. P. Williams, C. He, S. Martinson, M. Plomecka, L. Wei, Y. Zhou, Q.-Z. Zhu, M. Abraham, E. Brand, A. Bulanova, J. A. Cardille, C.… — [An AI system to help scientists write expert-level empirical software](https://md.research.elamur.ai/free/en/papers/aygun-2025-ai-system-help-scientists/)
40. D. A. Boiko, R. MacKnight, B. Kline, and G. Gomes. [Autonomous chemical research with large language models](https://md.research.elamur.ai/free/en/papers/boiko-2023-autonomous-chemical-research-large-language/). Nature, 624:570–578, 2023.
41. Y. Chen, C. Liu, Z. Chen, T. Liu, B. Han, and K. Zhang. CausalEvolve: Towards Open-Ended Discovery with Causal Scratchpad, Mar. 2026. [find](https://scholar.google.com/scholar?q=Y%2E+Chen%2C+C%2E+Liu%2C+Z%2E+Chen%2C+T%2E+Liu%2C+B%2E+Han%2C+and+K%2E+Zhang%2E+CausalEvolve%3A+Towards+Open%2DEnded+Discovery+with+Causal+Scratchpad%2C+Mar%2E+2026%2E)
42. M. Cobelli and S. Sanvito. Agentic design of compositional descriptors via autoresearch for materials science applications. arXiv preprint arXiv:2605.14671, 2026. — [arXiv:2605.14671](https://arxiv.org/abs/2605.14671) [find](https://scholar.google.com/scholar?q=M%2E+Cobelli+and+S%2E+Sanvito%2E+Agentic+design+of+compositional+descriptors+via+autoresearch+for+materials+science+applications%2E+arXiv+preprint+arXiv%3A2605%2E14671%2C+2026%2E+%E2%80%94+%5BarXiv%3A2605%2E14671%5D%28https%3A%2F%2Farxiv%2Eor)
43. L. Fan, P. Dai, Z. Deng, H. Wang, X. Gong, Y. Zheng, and Y. Ou. Evolving Medical Imaging Agents via Experience-driven Self-skill Discovery, Mar. 2026. arXiv:2603.05860 [cs]. — [arXiv:2603.05860](https://arxiv.org/abs/2603.05860) [find](https://scholar.google.com/scholar?q=L%2E+Fan%2C+P%2E+Dai%2C+Z%2E+Deng%2C+H%2E+Wang%2C+X%2E+Gong%2C+Y%2E+Zheng%2C+and+Y%2E+Ou%2E+Evolving+Medical+Imaging+Agents+via+Experience%2Ddriven+Self%2Dskill+Discovery%2C+Mar%2E+2026%2E+arXiv%3A2603%2E05860+%5Bcs%5D%2E+%E2%80%94+%5BarXiv%3A2603%2E05860%5D%28https)
44. F. Ferreira, L. Wobbe, A. Krishnakumar, F. Hutter, and A. Zela. Can LLMs Beat Classical Hyperparameter Optimization Algorithms? A Study on autoresearch, Mar. 2026. [find](https://scholar.google.com/scholar?q=F%2E+Ferreira%2C+L%2E+Wobbe%2C+A%2E+Krishnakumar%2C+F%2E+Hutter%2C+and+A%2E+Zela%2E+Can+LLMs+Beat+Classical+Hyperparameter+Optimization+Algorithms%3F+A+Study+on+autoresearch%2C+Mar%2E+2026%2E)
45. A. E. Ghareeb, B. Chang, L. Mitchener, A. Yiu, C. J. Szostkiewicz, D. Shved, G. J. Gyimesi, J. M. Laurent, S. M. Wright, M. T. Razzak, A. D. White, S. C. Finnemann, M. M. Hinks, and S. G. Rodriques. A multi-agent system for automating scientific discovery. Nature, May 2026. [find](https://scholar.google.com/scholar?q=A%2E+E%2E+Ghareeb%2C+B%2E+Chang%2C+L%2E+Mitchener%2C+A%2E+Yiu%2C+C%2E+J%2E+Szostkiewicz%2C+D%2E+Shved%2C+G%2E+J%2E+Gyimesi%2C+J%2E+M%2E+Laurent%2C+S%2E+M%2E+Wright%2C+M%2E+T%2E+Razzak%2C+A%2E+D%2E+White%2C+S%2E+C%2E+Finnemann%2C+M%2E+M%2E+Hinks%2C+and+S%2E+G%2E+Rodriques%2E+A)
46. J. Gottweis, W.-H. Weng, A. Daryin, T. Tu, P. Sirkovic, A. Myaskovsky, G. Glowaty, F. Weissenberger, A. Orlandi, D. Popovici, A. Palepu, K. Rong, R. Tanno, K. Saab, F. Zhang, J. Blum, A. Carroll, K. Kulkarni, N. Tomašev, D. Zverinski, I. Rendulic, E. Vedadi, F. Hasler, L. Rimanic, M. Boia, I. Budise… — [Accelerating scientific discovery with Co-Scientist](https://md.research.elamur.ai/free/en/papers/gottweis-2025-accelerating-scientific-discovery-co/)
47. T. Han, Y. Zhang, W. Song, C. Fang, Z. Chen, Y. Sun, and L. Hu. Swe-skills-bench: Do agent skills actually help in real-world software engineering? arXiv preprint arXiv:2603.15401, 2026. — [arXiv:2603.15401](https://arxiv.org/abs/2603.15401) [find](https://scholar.google.com/scholar?q=T%2E+Han%2C+Y%2E+Zhang%2C+W%2E+Song%2C+C%2E+Fang%2C+Z%2E+Chen%2C+Y%2E+Sun%2C+and+L%2E+Hu%2E+Swe%2Dskills%2Dbench%3A+Do+agent+skills+actually+help+in+real%2Dworld+software+engineering%3F+arXiv+preprint+arXiv%3A2603%2E15401%2C+2026%2E+%E2%80%94+%5BarXiv%3A2603)
48. C. He, X. Zhou, D. Wang, H. Xu, W. Liu, and C. Miao. The AutoResearch Moment: From Experimenter to Research Director, Mar. 2026. [find](https://scholar.google.com/scholar?q=C%2E+He%2C+X%2E+Zhou%2C+D%2E+Wang%2C+H%2E+Xu%2C+W%2E+Liu%2C+and+C%2E+Miao%2E+The+AutoResearch+Moment%3A+From+Experimenter+to+Research+Director%2C+Mar%2E+2026%2E)
49. M. He, F. Jiang, J. Jiao, M. Li, K. Li, Y. Liao, B. Liu, T. Liu, F. Qi, Z. Shang, W. Song, Y. Sun, X. Wang, H. Wang, D. Xiong, C. Yuan, B. Zhang, Z. Zhang, and X. Zhu. Dr.Sai: An agentic AI for real-world physics analysis at BESIII, Apr. 2026. arXiv:2604.22541 [hep-ex] version: 1. — [arXiv:2604.22541](https://arxiv.org/abs/2604.22541) [find](https://scholar.google.com/scholar?q=M%2E+He%2C+F%2E+Jiang%2C+J%2E+Jiao%2C+M%2E+Li%2C+K%2E+Li%2C+Y%2E+Liao%2C+B%2E+Liu%2C+T%2E+Liu%2C+F%2E+Qi%2C+Z%2E+Shang%2C+W%2E+Song%2C+Y%2E+Sun%2C+X%2E+Wang%2C+H%2E+Wang%2C+D%2E+Xiong%2C+C%2E+Yuan%2C+B%2E+Zhang%2C+Z%2E+Zhang%2C+and+X%2E+Zhu%2E+Dr%2ESai%3A+An+agentic+AI+for+real%2Dw)
50. Y. Huang, Y. Chen, H. Zhang, K. Li, H. Zhou, M. Fang, L. Yang, X. Li, L. Shang, S. Xu, J. Hao, K. Shao, and J. Wang. Deep Research Agents: A Systematic Examination And Roadmap, Sept. 2025. arXiv:2506.18096 [cs]. — [arXiv:2506.18096](https://arxiv.org/abs/2506.18096) [find](https://scholar.google.com/scholar?q=Y%2E+Huang%2C+Y%2E+Chen%2C+H%2E+Zhang%2C+K%2E+Li%2C+H%2E+Zhou%2C+M%2E+Fang%2C+L%2E+Yang%2C+X%2E+Li%2C+L%2E+Shang%2C+S%2E+Xu%2C+J%2E+Hao%2C+K%2E+Shao%2C+and+J%2E+Wang%2E+Deep+Research+Agents%3A+A+Systematic+Examination+And+Roadmap%2C+Sept%2E+2025%2E+arXiv%3A2506%2E)
51. V. Ilin. Semi-Autonomous Formalization of the Vlasov-Maxwell-Landau Equilibrium, Mar. 2026. [find](https://scholar.google.com/scholar?q=V%2E+Ilin%2E+Semi%2DAutonomous+Formalization+of+the+Vlasov%2DMaxwell%2DLandau+Equilibrium%2C+Mar%2E+2026%2E)
52. B. Jia, S. Kamboj, S. Katipomu, S. H. Han, N. Sengupta, and A. Jackson. Nomad: Autonomous Exploration and Discovery, Mar. 2026. [find](https://scholar.google.com/scholar?q=B%2E+Jia%2C+S%2E+Kamboj%2C+S%2E+Katipomu%2C+S%2E+H%2E+Han%2C+N%2E+Sengupta%2C+and+A%2E+Jackson%2E+Nomad%3A+Autonomous+Exploration+and+Discovery%2C+Mar%2E+2026%2E)
53. G. Jiang, Z. Su, X. Qu, and Y. R. Fung. Xskill: Continual learning from experience and skills in multimodal agents. arXiv preprint arXiv:2603.12056, 2026. — [arXiv:2603.12056](https://arxiv.org/abs/2603.12056) [find](https://scholar.google.com/scholar?q=G%2E+Jiang%2C+Z%2E+Su%2C+X%2E+Qu%2C+and+Y%2E+R%2E+Fung%2E+Xskill%3A+Continual+learning+from+experience+and+skills+in+multimodal+agents%2E+arXiv+preprint+arXiv%3A2603%2E12056%2C+2026%2E+%E2%80%94+%5BarXiv%3A2603%2E12056%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F26)
54. A. Karpathy. karpathy/autoresearch, Apr. 2026. [find](https://scholar.google.com/scholar?q=A%2E+Karpathy%2E+karpathy%2Fautoresearch%2C+Apr%2E+2026%2E)
55. J. Karwowski, O. Hayman, X. Bai, K. Kiendlhofer, C. Griffin, and J. M. V. Skalse. Goodhart’s law in reinforcement learning. In The Twelfth International Conference on Learning Representations, 2024. [find](https://scholar.google.com/scholar?q=J%2E+Karwowski%2C+O%2E+Hayman%2C+X%2E+Bai%2C+K%2E+Kiendlhofer%2C+C%2E+Griffin%2C+and+J%2E+M%2E+V%2E+Skalse%2E+Goodhart%E2%80%99s+law+in+reinforcement+learning%2E+In+The+Twelfth+International+Conference+on+Learning+Representations%2C+2024%2E)
56. N. Khandelwal and S. S. Gupta. Agent-Driven Autonomous Reinforcement Learning Research: Iterative Policy Improvement for Quadruped Locomotion, Mar. 2026. [find](https://scholar.google.com/scholar?q=N%2E+Khandelwal+and+S%2E+S%2E+Gupta%2E+Agent%2DDriven+Autonomous+Reinforcement+Learning+Research%3A+Iterative+Policy+Improvement+for+Quadruped+Locomotion%2C+Mar%2E+2026%2E)
57. L. Kong, X. Sun, W. Chow, L. Li, K. Q. Lin, X. B. Zhang, S. Wang, R. Li, Q. Wu, W. Gao, Y. Wang, S. Xie, J. Liu, L. Qu, S. Li, L. X. Ng, B. R. Cottereau, Z. Liu, T.-S. Chua, and W. T. Ooi. [AI for Auto-Research: Roadmap & user guide](https://md.research.elamur.ai/free/en/papers/kong-2026-ai-auto-research-roadmap-user/). May 2026.
58. S. Kuroki, T. Nakamura, T. Akiba, and Y. Tang. Agent skill acquisition for large language models via cycleqd. arXiv preprint arXiv:2410.14735, 2024. — [arXiv:2410.14735](https://arxiv.org/abs/2410.14735) [find](https://scholar.google.com/scholar?q=S%2E+Kuroki%2C+T%2E+Nakamura%2C+T%2E+Akiba%2C+and+Y%2E+Tang%2E+Agent+skill+acquisition+for+large+language+models+via+cycleqd%2E+arXiv+preprint+arXiv%3A2410%2E14735%2C+2024%2E+%E2%80%94+%5BarXiv%3A2410%2E14735%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2410%2E147)
59. C.-Y. Lee, H. Liang, R. Kim, A. McDannald, C. A. R. Ocampo, A. G. Kusne, and I. Takeuchi. Real-time multi-instrument autonomous discovery of novel phase-change memory materials. May 2026. [find](https://scholar.google.com/scholar?q=C%2E%2DY%2E+Lee%2C+H%2E+Liang%2C+R%2E+Kim%2C+A%2E+McDannald%2C+C%2E+A%2E+R%2E+Ocampo%2C+A%2E+G%2E+Kusne%2C+and+I%2E+Takeuchi%2E+Real%2Dtime+multi%2Dinstrument+autonomous+discovery+of+novel+phase%2Dchange+memory+materials%2E+May+2026%2E)
60. F. Li, P. Tagkopoulos, and I. Tagkopoulos. Skillflow: Scalable and efficient agent skill retrieval system. arXiv e-prints, pages arXiv–2504, 2025. [find](https://scholar.google.com/scholar?q=F%2E+Li%2C+P%2E+Tagkopoulos%2C+and+I%2E+Tagkopoulos%2E+Skillflow%3A+Scalable+and+efficient+agent+skill+retrieval+system%2E+arXiv+e%2Dprints%2C+pages+arXiv%E2%80%932504%2C+2025%2E)
61. H. Li, C. Mu, J. Chen, S. Ren, Z. Cui, Y. Zhang, L. Bai, and S. Hu. Organizing, orchestrating, and benchmarking agent skills at ecosystem scale. arXiv preprint arXiv:2603.02176, 2026. — [arXiv:2603.02176](https://arxiv.org/abs/2603.02176) [find](https://scholar.google.com/scholar?q=H%2E+Li%2C+C%2E+Mu%2C+J%2E+Chen%2C+S%2E+Ren%2C+Z%2E+Cui%2C+Y%2E+Zhang%2C+L%2E+Bai%2C+and+S%2E+Hu%2E+Organizing%2C+orchestrating%2C+and+benchmarking+agent+skills+at+ecosystem+scale%2E+arXiv+preprint+arXiv%3A2603%2E02176%2C+2026%2E+%E2%80%94+%5BarXiv%3A2603%2E02)
62. X. Li. Auto Researching, not hyperparameter tuning: Convergence Analysis of 10,000 Experiments, Mar. 2026. [find](https://scholar.google.com/scholar?q=X%2E+Li%2E+Auto+Researching%2C+not+hyperparameter+tuning%3A+Convergence+Analysis+of+10%2C000+Experiments%2C+Mar%2E+2026%2E)
63. X. Li. When single-agent with skills replace multi-agent systems and when they fail. arXiv preprint arXiv:2601.04748, 2026. — [arXiv:2601.04748](https://arxiv.org/abs/2601.04748) [find](https://scholar.google.com/scholar?q=X%2E+Li%2E+When+single%2Dagent+with+skills+replace+multi%2Dagent+systems+and+when+they+fail%2E+arXiv+preprint+arXiv%3A2601%2E04748%2C+2026%2E+%E2%80%94+%5BarXiv%3A2601%2E04748%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2601%2E04748%29)
64. X. Li, W. Chen, Y. Liu, S. Zheng, X. Chen, Y. He, Y. Li, B. You, H. Shen, J. Sun, et al. Skillsbench: Benchmarking how well agent skills work across diverse tasks. arXiv preprint arXiv:2602.12670, 2026. — [arXiv:2602.12670](https://arxiv.org/abs/2602.12670) [find](https://scholar.google.com/scholar?q=X%2E+Li%2C+W%2E+Chen%2C+Y%2E+Liu%2C+S%2E+Zheng%2C+X%2E+Chen%2C+Y%2E+He%2C+Y%2E+Li%2C+B%2E+You%2C+H%2E+Shen%2C+J%2E+Sun%2C+et+al%2E+Skillsbench%3A+Benchmarking+how+well+agent+skills+work+across+diverse+tasks%2E+arXiv+preprint+arXiv%3A2602%2E12670%2C+202)
65. Y. Li, C. Shao, X. Liu, R. Zhao, P. Liu, H. Su, Z. Chen, Q. Yang, A. Xu, Y. Fang, et al. [Autosota: An end-to-end automated research system for state-of-the-art ai model discovery](https://md.research.elamur.ai/free/en/papers/li-2026-autosota-end-end-automated-research/). arXiv preprint arXiv:2604.05550, 2026.
66. Y. Liang, R. Zhong, H. Xu, C. Jiang, Y. Zhong, R. Fang, J.-C. Gu, S. Deng, Y. Yao, M. Wang, et al. Skillnet: Create, evaluate, and connect ai skills. arXiv preprint arXiv:2603.04448, 2026. — [arXiv:2603.04448](https://arxiv.org/abs/2603.04448) [find](https://scholar.google.com/scholar?q=Y%2E+Liang%2C+R%2E+Zhong%2C+H%2E+Xu%2C+C%2E+Jiang%2C+Y%2E+Zhong%2C+R%2E+Fang%2C+J%2E%2DC%2E+Gu%2C+S%2E+Deng%2C+Y%2E+Yao%2C+M%2E+Wang%2C+et+al%2E+Skillnet%3A+Create%2C+evaluate%2C+and+connect+ai+skills%2E+arXiv+preprint+arXiv%3A2603%2E04448%2C+2026%2E+%E2%80%94+%5BarXiv%3A26)
67. C. Liu, T. Li, M. Huang, X. Wei, P. Liu, Y. Shen, Y. Mao, and T. Cui. Protrlsearch: A multi-round multimodal protein search agent with large language models trained via reinforcement learning. arXiv preprint arXiv:2603.01464, 2026. — [arXiv:2603.01464](https://arxiv.org/abs/2603.01464) [find](https://scholar.google.com/scholar?q=C%2E+Liu%2C+T%2E+Li%2C+M%2E+Huang%2C+X%2E+Wei%2C+P%2E+Liu%2C+Y%2E+Shen%2C+Y%2E+Mao%2C+and+T%2E+Cui%2E+Protrlsearch%3A+A+multi%2Dround+multimodal+protein+search+agent+with+large+language+models+trained+via+reinforcement+learning%2E+arXiv+p)
68. F. Liu, J. Han, T. Lyu, W. Zhang, Z.-R. Yang, L. Dai, C. Liu, and H. Liu. Foundation models for scientific discovery: From paradigm enhancement to paradigm transition. Advances in Neural Information Processing Systems, 2025. [find](https://scholar.google.com/scholar?q=F%2E+Liu%2C+J%2E+Han%2C+T%2E+Lyu%2C+W%2E+Zhang%2C+Z%2E%2DR%2E+Yang%2C+L%2E+Dai%2C+C%2E+Liu%2C+and+H%2E+Liu%2E+Foundation+models+for+scientific+discovery%3A+From+paradigm+enhancement+to+paradigm+transition%2E+Advances+in+Neural+Information+P)
69. F. Liu, J. Xu, X. Cui, X. Wang, Z. Guo, J. Wang, S. M. Mousavi, X. Gu, H. Chen, B. Fei, L. Fang, F. Ling, Z. Li, and L. Bai. TRACE: A Multi-Agent System for Autonomous Physical Reasoning for Seismology, Mar. 2026. [find](https://scholar.google.com/scholar?q=F%2E+Liu%2C+J%2E+Xu%2C+X%2E+Cui%2C+X%2E+Wang%2C+Z%2E+Guo%2C+J%2E+Wang%2C+S%2E+M%2E+Mousavi%2C+X%2E+Gu%2C+H%2E+Chen%2C+B%2E+Fei%2C+L%2E+Fang%2C+F%2E+Ling%2C+Z%2E+Li%2C+and+L%2E+Bai%2E+TRACE%3A+A+Multi%2DAgent+System+for+Autonomous+Physical+Reasoning+for+Seismolog)
70. J. Liu, Z. Ling, S. Qiu, Y. Liu, S. Han, P. Xia, H. Tu, Z. Zheng, C. Xie, C. Fleming, et al. Omni-simplemem: Autoresearch-guided discovery of lifelong multimodal agent memory. arXiv e-prints, pages arXiv–2604, 2026. [find](https://scholar.google.com/scholar?q=J%2E+Liu%2C+Z%2E+Ling%2C+S%2E+Qiu%2C+Y%2E+Liu%2C+S%2E+Han%2C+P%2E+Xia%2C+H%2E+Tu%2C+Z%2E+Zheng%2C+C%2E+Xie%2C+C%2E+Fleming%2C+et+al%2E+Omni%2Dsimplemem%3A+Autoresearch%2Dguided+discovery+of+lifelong+multimodal+agent+memory%2E+arXiv+e%2Dprints%2C+pages+ar)
71. J. Liu, S. Qiu, M. Li, B. Li, H. Ji, S. Han, X. Ye, P. Xia, Z. Dong, C. Zhang, L. Zhang, G. Chen, H. Tu, X. Yang, L. Feng, X. Zhao, H. Chen, J. Zhou, X. Wang, W. Zhang, H. Zhu, Y. Li, J. Mei, H. Fei, J. Zhang, L. Li, L. Zhang, Y. Zhou, S. Wang, C. Xiong, J. Zou, Z. Zheng, C. Xie, M. Ding, and H. Yao… [find](https://scholar.google.com/scholar?q=J%2E+Liu%2C+S%2E+Qiu%2C+M%2E+Li%2C+B%2E+Li%2C+H%2E+Ji%2C+S%2E+Han%2C+X%2E+Ye%2C+P%2E+Xia%2C+Z%2E+Dong%2C+C%2E+Zhang%2C+L%2E+Zhang%2C+G%2E+Chen%2C+H%2E+Tu%2C+X%2E+Yang%2C+L%2E+Feng%2C+X%2E+Zhao%2C+H%2E+Chen%2C+J%2E+Zhou%2C+X%2E+Wang%2C+W%2E+Zhang%2C+H%2E+Zhu%2C+Y%2E+Li%2C+J%2E+Mei%2C+H%2E+Fei%2C+)
72. J. Liu, J. Shen, S. Song, T. Li, X. Liu, R. Li, Z. Huang, J. Lin, J. Ning, C. Ji, S. Luo, W. Li, C. Ma, M. Hu, J. Xiong, J. Ye, B. Fu, N. Xu, Y. Chen, L. Jin, H. Chen, and J. He. MedProbeBench: Systematic Benchmarking at Deep Evidence Integration for Expert-level Medical Guideline, Apr. 2026. arXiv:… — [arXiv:2604.18418](https://arxiv.org/abs/2604.18418) [find](https://scholar.google.com/scholar?q=J%2E+Liu%2C+J%2E+Shen%2C+S%2E+Song%2C+T%2E+Li%2C+X%2E+Liu%2C+R%2E+Li%2C+Z%2E+Huang%2C+J%2E+Lin%2C+J%2E+Ning%2C+C%2E+Ji%2C+S%2E+Luo%2C+W%2E+Li%2C+C%2E+Ma%2C+M%2E+Hu%2C+J%2E+Xiong%2C+J%2E+Ye%2C+B%2E+Fu%2C+N%2E+Xu%2C+Y%2E+Chen%2C+L%2E+Jin%2C+H%2E+Chen%2C+and+J%2E+He%2E+MedProbeBench%3A+System)
73. J. Liu, X. Ye, P. Xia, Z. Zheng, C. Xie, M. Ding, and H. Yao. Evolvemem: Self-evolving memory architecture via autoresearch for llm agents. arXiv preprint arXiv:2605.13941, 2026. — [arXiv:2605.13941](https://arxiv.org/abs/2605.13941) [find](https://scholar.google.com/scholar?q=J%2E+Liu%2C+X%2E+Ye%2C+P%2E+Xia%2C+Z%2E+Zheng%2C+C%2E+Xie%2C+M%2E+Ding%2C+and+H%2E+Yao%2E+Evolvemem%3A+Self%2Devolving+memory+architecture+via+autoresearch+for+llm+agents%2E+arXiv+preprint+arXiv%3A2605%2E13941%2C+2026%2E+%E2%80%94+%5BarXiv%3A2605%2E13941%5D%28)
74. C. Lu, C. Lu, R. T. Lange, Y. Yamada, S. Hu, J. Foerster, D. Ha, and J. Clune. Towards end-to-end automation of AI research. Nature, 651(8107):914–919, Mar. 2026. [find](https://scholar.google.com/scholar?q=C%2E+Lu%2C+C%2E+Lu%2C+R%2E+T%2E+Lange%2C+Y%2E+Yamada%2C+S%2E+Hu%2C+J%2E+Foerster%2C+D%2E+Ha%2C+and+J%2E+Clune%2E+Towards+end%2Dto%2Dend+automation+of+AI+research%2E+Nature%2C+651%288107%29%3A914%E2%80%93919%2C+Mar%2E+2026%2E)
75. D. Manheim and S. Garrabrant. Categorizing variants of goodhart’s law. arXiv preprint arXiv:1803.04585, 2018. — [arXiv:1803.04585](https://arxiv.org/abs/1803.04585) [find](https://scholar.google.com/scholar?q=D%2E+Manheim+and+S%2E+Garrabrant%2E+Categorizing+variants+of+goodhart%E2%80%99s+law%2E+arXiv+preprint+arXiv%3A1803%2E04585%2C+2018%2E+%E2%80%94+%5BarXiv%3A1803%2E04585%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1803%2E04585%29)
76. L. Messeri and M. J. Crockett. Artificial intelligence and illusions of understanding in scientific research. Nature, 627:49–58, 2024. [find](https://scholar.google.com/scholar?q=L%2E+Messeri+and+M%2E+J%2E+Crockett%2E+Artificial+intelligence+and+illusions+of+understanding+in+scientific+research%2E+Nature%2C+627%3A49%E2%80%9358%2C+2024%2E)
77. J. Ni, Y. Liu, X. Liu, Y. Sun, M. Zhou, P. Cheng, D. Wang, X. Jiang, and G. Jiang. [Trace2skill: Distill trajectory-local lessons into transferable agent skills](https://md.research.elamur.ai/free/en/papers/ni-2026-trace2skill-distill-trajectory-local-lessons/). arXiv preprint arXiv:2603.25158, 2026.
78. A. Novikov, M. Balog, M. P. Kumar, et al. [AlphaEvolve: A coding agent for scientific and algorithmic discovery](https://md.research.elamur.ai/free/en/papers/novikov-2025-alphaevolve-coding-agent-scientific-algorithmic/), 2025.
79. OpenAI. Introducing deep research, 2025. [find](https://scholar.google.com/scholar?q=OpenAI%2E+Introducing+deep+research%2C+2025%2E)
80. A. Pan, E. Jones, M. Jagadeesan, and J. Steinhardt. [Feedback loops with language models drive in-context reward hacking](https://md.research.elamur.ai/free/en/papers/pan-2024-feedback-loops-language-models-drive/). In Proceedings of the 41st International Conference on Machine Learning, volume 235 of Proceedings of Machine Learning Research, pages 39154–39200. PMLR, 2024.
81. A. Panfilov, P. Romov, I. Shilov, Y.-A. d. Montjoye, J. Geiping, and M. Andriushchenko. Claudini: Autoresearch Discovers State-of-the-Art Adversarial Attack Algorithms for LLMs, Mar. 2026. [find](https://scholar.google.com/scholar?q=A%2E+Panfilov%2C+P%2E+Romov%2C+I%2E+Shilov%2C+Y%2E%2DA%2E+d%2E+Montjoye%2C+J%2E+Geiping%2C+and+M%2E+Andriushchenko%2E+Claudini%3A+Autoresearch+Discovers+State%2Dof%2Dthe%2DArt+Adversarial+Attack+Algorithms+for+LLMs%2C+Mar%2E+2026%2E)
82. A. Pepe, C.-Y. Lin, D. Magka, B. Acun, Y. N. Wu, A. Protopopov, C.-J. Wu, and Y. Bachrach. Agentic discovery of neural architectures: AIRA-Compose and AIRA-Design. May 2026. [find](https://scholar.google.com/scholar?q=A%2E+Pepe%2C+C%2E%2DY%2E+Lin%2C+D%2E+Magka%2C+B%2E+Acun%2C+Y%2E+N%2E+Wu%2C+A%2E+Protopopov%2C+C%2E%2DJ%2E+Wu%2C+and+Y%2E+Bachrach%2E+Agentic+discovery+of+neural+architectures%3A+AIRA%2DCompose+and+AIRA%2DDesign%2E+May+2026%2E)
83. Y. Qian, X. Hu, J. Yu, S. Xin, X. Chen, J. Zhang, P.-T. Jiang, J. Liu, and H. B. Li. Medmaslab: A unified orchestration framework for benchmarking multimodal medical multi-agent systems. arXiv preprint arXiv:2603.09909, 2026. — [arXiv:2603.09909](https://arxiv.org/abs/2603.09909) [find](https://scholar.google.com/scholar?q=Y%2E+Qian%2C+X%2E+Hu%2C+J%2E+Yu%2C+S%2E+Xin%2C+X%2E+Chen%2C+J%2E+Zhang%2C+P%2E%2DT%2E+Jiang%2C+J%2E+Liu%2C+and+H%2E+B%2E+Li%2E+Medmaslab%3A+A+unified+orchestration+framework+for+benchmarking+multimodal+medical+multi%2Dagent+systems%2E+arXiv+preprin)
84. J. Qiao, W. Meng, Y. Cheng, Z. Lin, Z. Zhang, X. Tan, J. Gong, K. Shao, and Y. Xie. Memory intelligence agent. arXiv preprint arXiv:2604.04503, 2026. — [arXiv:2604.04503](https://arxiv.org/abs/2604.04503) [find](https://scholar.google.com/scholar?q=J%2E+Qiao%2C+W%2E+Meng%2C+Y%2E+Cheng%2C+Z%2E+Lin%2C+Z%2E+Zhang%2C+X%2E+Tan%2C+J%2E+Gong%2C+K%2E+Shao%2C+and+Y%2E+Xie%2E+Memory+intelligence+agent%2E+arXiv+preprint+arXiv%3A2604%2E04503%2C+2026%2E+%E2%80%94+%5BarXiv%3A2604%2E04503%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2604%2E04)
85. Y. Qin, Y. Yan, J. Pang, and X. Zhang. BloClaw: An Omniscient, Multi-Modal Agentic Workspace for Next-Generation Scientific Discovery, Apr. 2026. [find](https://scholar.google.com/scholar?q=Y%2E+Qin%2C+Y%2E+Yan%2C+J%2E+Pang%2C+and+X%2E+Zhang%2E+BloClaw%3A+An+Omniscient%2C+Multi%2DModal+Agentic+Workspace+for+Next%2DGeneration+Scientific+Discovery%2C+Apr%2E+2026%2E)
86. A. Qu, H. Zheng, Z. Zhou, Y. Yan, Y. Tang, S. Y. Ong, F. Hong, K. Zhou, C. Jiang, M. Kong, et al. [Coral: Towards autonomous multi-agent evolution for open-ended discovery](https://md.research.elamur.ai/free/en/papers/qu-2026-coral-towards-autonomous-multi-agent/). arXiv preprint arXiv:2604.01658, 2026.
87. Y. Qu, K. Huang, M. Yin, K. Zhan, D. Liu, D. Yin, H. C. Cousins, W. A. Johnson, X. Wang, M. Shah, et al. [Crispr-gpt for agentic automation of gene-editing experiments](https://md.research.elamur.ai/free/en/papers/qu-2024-crispr-gpt-agentic-automation/). Nature Biomedical Engineering, 10(2):245–258, 2026.
88. Y. Qu and M. Lu. Bilevel Autoresearch: Meta-Autoresearching Itself, Mar. 2026. [find](https://scholar.google.com/scholar?q=Y%2E+Qu+and+M%2E+Lu%2E+Bilevel+Autoresearch%3A+Meta%2DAutoresearching+Itself%2C+Mar%2E+2026%2E)
89. M. Ríos-García, N. Alampara, C. Gupta, I. Mandal, S. Mannan, A. A. Aghajani, N. Krishnan, and K. M. Jablonka. Ai scientists produce results without reasoning scientifically. arXiv preprint arXiv:2604.18805, 2026. [find](https://scholar.google.com/scholar?q=M%2E+R%C3%ADos%2DGarc%C3%ADa%2C+N%2E+Alampara%2C+C%2E+Gupta%2C+I%2E+Mandal%2C+S%2E+Mannan%2C+A%2E+A%2E+Aghajani%2C+N%2E+Krishnan%2C+and+K%2E+M%2E+Jablonka%2E+Ai+scientists+produce+results+without+reasoning+scientifically%2E+arXiv+preprint+arXiv%3A2604%2E)
90. B. Romera-Paredes, M. Barekatain, A. Novikov, M. Balog, M. P. Kumar, E. Dupont, F. J. R. Ruiz, J. S. Ellenberg, P. Wang, O. Fawzi, P. Kohli, and A. Fawzi. [Mathematical discoveries from program search with large language models](https://md.research.elamur.ai/free/en/papers/paredes-2023-mathematical-discoveries-program-search-large/). Nature, 625:468–475, 2024.
91. S. Schmidgall and M. Moor. [AgentRxiv: Towards Collaborative Autonomous Research](https://md.research.elamur.ai/free/en/papers/schmidgall-2025-agentrxiv-collaborative-autonomous-research/). arXiv preprint arXiv:2503.18102, Mar. 2025. Companion paper to Agent Laboratory repository.
92. W. Shen, Y. Hu, C. Liu, J. Wu, J. Zhu, C. Shen, M. Xu, Y. Jin, B. Wiestler, D. Rueckert, and J. Pan. MedOpenClaw: Auditable Medical Imaging Agents Reasoning over Uncurated Full Studies, Mar. 2026. [find](https://scholar.google.com/scholar?q=W%2E+Shen%2C+Y%2E+Hu%2C+C%2E+Liu%2C+J%2E+Wu%2C+J%2E+Zhu%2C+C%2E+Shen%2C+M%2E+Xu%2C+Y%2E+Jin%2C+B%2E+Wiestler%2C+D%2E+Rueckert%2C+and+J%2E+Pan%2E+MedOpenClaw%3A+Auditable+Medical+Imaging+Agents+Reasoning+over+Uncurated+Full+Studies%2C+Mar%2E+2026%2E)
93. Y. Shen, Z. Yi, Z. Zhao, L. Sun, D. Li, C.-T. Lin, and Y. Shi. An Empirical Study of Multi-Agent Collaboration for Automated Research, Mar. 2026. [find](https://scholar.google.com/scholar?q=Y%2E+Shen%2C+Z%2E+Yi%2C+Z%2E+Zhao%2C+L%2E+Sun%2C+D%2E+Li%2C+C%2E%2DT%2E+Lin%2C+and+Y%2E+Shi%2E+An+Empirical+Study+of+Multi%2DAgent+Collaboration+for+Automated+Research%2C+Mar%2E+2026%2E)
94. L. Shi, Z. J. Zheng, X. Juan, Y. Wang, M. Yin, M. Sengupta, K. Wolinski, Y. Jia, J. Shi, D. Saucedo, N. Saggi, H. Guan, K. Watanabe, T. Taniguchi, A. Yazdani, M. Wang, and S. Wu. Qumus: Realization of an embodied AI quantum material experimentalist. May 2026. [find](https://scholar.google.com/scholar?q=L%2E+Shi%2C+Z%2E+J%2E+Zheng%2C+X%2E+Juan%2C+Y%2E+Wang%2C+M%2E+Yin%2C+M%2E+Sengupta%2C+K%2E+Wolinski%2C+Y%2E+Jia%2C+J%2E+Shi%2C+D%2E+Saucedo%2C+N%2E+Saggi%2C+H%2E+Guan%2C+K%2E+Watanabe%2C+T%2E+Taniguchi%2C+A%2E+Yazdani%2C+M%2E+Wang%2C+and+S%2E+Wu%2E+Qumus%3A+Realization+of)
95. N. Somasekharan, R. Pathak, M. Dhanakoti, T. Zhang, L. Yue, A. Zhu, and S. Pan. Ai cfd scientist: Toward open-ended computational fluid dynamics discovery with physics-aware ai agents. arXiv preprint arXiv:2605.06607, 2026. — [arXiv:2605.06607](https://arxiv.org/abs/2605.06607) [find](https://scholar.google.com/scholar?q=N%2E+Somasekharan%2C+R%2E+Pathak%2C+M%2E+Dhanakoti%2C+T%2E+Zhang%2C+L%2E+Yue%2C+A%2E+Zhu%2C+and+S%2E+Pan%2E+Ai+cfd+scientist%3A+Toward+open%2Dended+computational+fluid+dynamics+discovery+with+physics%2Daware+ai+agents%2E+arXiv+preprint+)
96. W. Song, J. Yue, and Z. Pang. Abstral: Automatic design of multi-agent systems through iterative refinement and topology optimization. arXiv preprint arXiv:2603.22791, 2026. — [arXiv:2603.22791](https://arxiv.org/abs/2603.22791) [find](https://scholar.google.com/scholar?q=W%2E+Song%2C+J%2E+Yue%2C+and+Z%2E+Pang%2E+Abstral%3A+Automatic+design+of+multi%2Dagent+systems+through+iterative+refinement+and+topology+optimization%2E+arXiv+preprint+arXiv%3A2603%2E22791%2C+2026%2E+%E2%80%94+%5BarXiv%3A2603%2E22791%5D%28https)
97. J. Su, S. Tu, B. Sun, and X. Liang. STRIDE: A self-reflective agent framework for reliable automatic equation discovery. May 2026. [find](https://scholar.google.com/scholar?q=J%2E+Su%2C+S%2E+Tu%2C+B%2E+Sun%2C+and+X%2E+Liang%2E+STRIDE%3A+A+self%2Dreflective+agent+framework+for+reliable+automatic+equation+discovery%2E+May+2026%2E)
98. Z. Sun, Z. Liu, Y. Zang, Y. Cao, X. Dong, T. Wu, D. Lin, and J. Wang. Seagent: Self-evolving computer use agent with autonomous learning from experience. arXiv preprint arXiv:2508.04700, 2025. — [arXiv:2508.04700](https://arxiv.org/abs/2508.04700) [find](https://scholar.google.com/scholar?q=Z%2E+Sun%2C+Z%2E+Liu%2C+Y%2E+Zang%2C+Y%2E+Cao%2C+X%2E+Dong%2C+T%2E+Wu%2C+D%2E+Lin%2C+and+J%2E+Wang%2E+Seagent%3A+Self%2Devolving+computer+use+agent+with+autonomous+learning+from+experience%2E+arXiv+preprint+arXiv%3A2508%2E04700%2C+2025%2E+%E2%80%94+%5BarXi)
99. L. L. Team, S. M. Schmon, D. Pretorius, S. Mathis, R. Bartke-Croughan, A. Puvanendran, J. Vuckovic, H. Kenlay, M. Vlachynská, A. Bridgland, I. Grishin, S. Over, D. Li, B. Li, J. Crabbé, A. Hilmkil, A. W. R. Nelson, D. Yuan, A. Obika, and S. A. A. Kohl. Latent-Y: A Lab-Validated Autonomous Agent for… [find](https://scholar.google.com/scholar?q=L%2E+L%2E+Team%2C+S%2E+M%2E+Schmon%2C+D%2E+Pretorius%2C+S%2E+Mathis%2C+R%2E+Bartke%2DCroughan%2C+A%2E+Puvanendran%2C+J%2E+Vuckovic%2C+H%2E+Kenlay%2C+M%2E+Vlachynsk%C3%A1%2C+A%2E+Bridgland%2C+I%2E+Grishin%2C+S%2E+Over%2C+D%2E+Li%2C+B%2E+Li%2C+J%2E+Crabb%C3%A9%2C+A%2E+Hilmkil%2C+A%2E)
100. M. Team, S. Bai, L. Bing, L. Lei, R. Li, X. Li, X. Lin, E. Min, L. Su, B. Wang, L. Wang, L. Wang, S. Wang, X. Wang, Y. Zhang, Z. Zhang, G. Chen, L. Chen, Z. Cheng, Y. Deng, Z. Huang, D. Ng, J. Ni, Q. Ren, X. Tang, B. L. Wang, H. Wang, N. Wang, C. Wei, Q. Wu, J. Xia, Y. Xiao, H. Xu, X. Xu, C. Xue, Z.… [find](https://scholar.google.com/scholar?q=M%2E+Team%2C+S%2E+Bai%2C+L%2E+Bing%2C+L%2E+Lei%2C+R%2E+Li%2C+X%2E+Li%2C+X%2E+Lin%2C+E%2E+Min%2C+L%2E+Su%2C+B%2E+Wang%2C+L%2E+Wang%2C+L%2E+Wang%2C+S%2E+Wang%2C+X%2E+Wang%2C+Y%2E+Zhang%2C+Z%2E+Zhang%2C+G%2E+Chen%2C+L%2E+Chen%2C+Z%2E+Cheng%2C+Y%2E+Deng%2C+Z%2E+Huang%2C+D%2E+Ng%2C+J%2E+Ni%2C+Q%2E+)
101. J. D. Toscano, Z. Chai, and G. E. Karniadakis. GRAFT-ATHENA: Self-improving agentic teams for autonomous discovery and evolutionary numerical algorithms. May 2026. [find](https://scholar.google.com/scholar?q=J%2E+D%2E+Toscano%2C+Z%2E+Chai%2C+and+G%2E+E%2E+Karniadakis%2E+GRAFT%2DATHENA%3A+Self%2Dimproving+agentic+teams+for+autonomous+discovery+and+evolutionary+numerical+algorithms%2E+May+2026%2E)
102. J. Wang, Z. Tao, H. Zeng, Z. Yang, H. Zamani, and H. Yu. Tarse: Test-time adaptation via retrieval of skills and experience for reasoning agents. arXiv preprint arXiv:2603.01241, 2026. — [arXiv:2603.01241](https://arxiv.org/abs/2603.01241) [find](https://scholar.google.com/scholar?q=J%2E+Wang%2C+Z%2E+Tao%2C+H%2E+Zeng%2C+Z%2E+Yang%2C+H%2E+Zamani%2C+and+H%2E+Yu%2E+Tarse%3A+Test%2Dtime+adaptation+via+retrieval+of+skills+and+experience+for+reasoning+agents%2E+arXiv+preprint+arXiv%3A2603%2E01241%2C+2026%2E+%E2%80%94+%5BarXiv%3A2603%2E0)
103. L. Wang, Y. He, P. Chen, A. Yehudai, Y. Liu, R. Ying, M. Shmueli-Scheuer, and A. Cohan. Time to REFLECT: Can we trust LLM judges for evidence-based research agents? 2026. [find](https://scholar.google.com/scholar?q=L%2E+Wang%2C+Y%2E+He%2C+P%2E+Chen%2C+A%2E+Yehudai%2C+Y%2E+Liu%2C+R%2E+Ying%2C+M%2E+Shmueli%2DScheuer%2C+and+A%2E+Cohan%2E+Time+to+REFLECT%3A+Can+we+trust+LLM+judges+for+evidence%2Dbased+research+agents%3F+2026%2E)
104. L. Wang, Y. Li, J. Wu, H. Gao, X. Bo, X. Chen, and J.-R. Wen. Llm agents as social scientists: A human-ai collaborative platform for social science automation. arXiv preprint arXiv:2604.01520, 2026. — [arXiv:2604.01520](https://arxiv.org/abs/2604.01520) [find](https://scholar.google.com/scholar?q=L%2E+Wang%2C+Y%2E+Li%2C+J%2E+Wu%2C+H%2E+Gao%2C+X%2E+Bo%2C+X%2E+Chen%2C+and+J%2E%2DR%2E+Wen%2E+Llm+agents+as+social+scientists%3A+A+human%2Dai+collaborative+platform+for+social+science+automation%2E+arXiv+preprint+arXiv%3A2604%2E01520%2C+2026%2E+%E2%80%94)
105. S. Wang, J. Jin, R. Fu, Z. Yan, X. Wang, M. Hu, E. Wang, X. Li, K. Zhang, L. Yao, W. Jiao, X. Cheng, Y. Lu, and Z. Ge. MuSEAgent: A Multimodal Reasoning Agent with Stateful Experiences, Mar. 2026. [find](https://scholar.google.com/scholar?q=S%2E+Wang%2C+J%2E+Jin%2C+R%2E+Fu%2C+Z%2E+Yan%2C+X%2E+Wang%2C+M%2E+Hu%2C+E%2E+Wang%2C+X%2E+Li%2C+K%2E+Zhang%2C+L%2E+Yao%2C+W%2E+Jiao%2C+X%2E+Cheng%2C+Y%2E+Lu%2C+and+Z%2E+Ge%2E+MuSEAgent%3A+A+Multimodal+Reasoning+Agent+with+Stateful+Experiences%2C+Mar%2E+2026%2E)
106. Y. Wang, G. Sinenka, and X. Brace. Rhizome os-1: Rhizome’s semi-autonomous operating system for small molecule drug discovery. arXiv preprint arXiv:2604.07512, 2026. — [arXiv:2604.07512](https://arxiv.org/abs/2604.07512) [find](https://scholar.google.com/scholar?q=Y%2E+Wang%2C+G%2E+Sinenka%2C+and+X%2E+Brace%2E+Rhizome+os%2D1%3A+Rhizome%E2%80%99s+semi%2Dautonomous+operating+system+for+small+molecule+drug+discovery%2E+arXiv+preprint+arXiv%3A2604%2E07512%2C+2026%2E+%E2%80%94+%5BarXiv%3A2604%2E07512%5D%28https%3A%2F%2Farxiv)
107. J. Wei, Y. Yang, X. Zhang, Y. Chen, X. Zhuang, Z. Gao, D. Zhou, G. Wang, Z. Gao, J. Cao, Z. Qiu, M. Hu, C. Ma, S. Tang, J. He, C. Song, X. He, Q. Zhang, C. You, S. Zheng, N. Ding, W. Ouyang, N. Dong, Y. Cheng, S. Sun, L. Bai, and B. Zhou. From AI for Science to Agentic Science: A Survey on Autonomou… — [From AI for Science to Agentic Science: A Survey on Autonomous Scientific Discovery](https://md.research.elamur.ai/free/en/papers/wei-2025-ai-science-agentic-science/)
108. Y. Weng, M. Zhu, G. Bao, H. Zhang, J. Wang, Y. Zhang, and L. Yang. [Cycleresearcher: Improving automated research via automated review](https://md.research.elamur.ai/free/en/papers/weng-2024-cycleresearcher-improving-automated-research/). In The Thirteenth International Conference on Learning Representations, 2025.
109. H. Wu, B. Zheng, D. Song, Y. Jiang, J. Gao, L. Xing, L. Sun, and Y. Yuan. Towards a Medical AI Scientist, Mar. 2026. [find](https://scholar.google.com/scholar?q=H%2E+Wu%2C+B%2E+Zheng%2C+D%2E+Song%2C+Y%2E+Jiang%2C+J%2E+Gao%2C+L%2E+Xing%2C+L%2E+Sun%2C+and+Y%2E+Yuan%2E+Towards+a+Medical+AI+Scientist%2C+Mar%2E+2026%2E)
110. P. Xia, J. Chen, H. Wang, J. Liu, K. Zeng, Y. Wang, S. Han, Y. Zhou, X. Zhao, H. Chen, et al. Skillrl: Evolving agents via recursive skill-augmented reinforcement learning. arXiv preprint arXiv:2602.08234, 2026. — [arXiv:2602.08234](https://arxiv.org/abs/2602.08234) [find](https://scholar.google.com/scholar?q=P%2E+Xia%2C+J%2E+Chen%2C+H%2E+Wang%2C+J%2E+Liu%2C+K%2E+Zeng%2C+Y%2E+Wang%2C+S%2E+Han%2C+Y%2E+Zhou%2C+X%2E+Zhao%2C+H%2E+Chen%2C+et+al%2E+Skillrl%3A+Evolving+agents+via+recursive+skill%2Daugmented+reinforcement+learning%2E+arXiv+preprint+arXiv%3A2602%2E0)
111. S. Xia, Y. Zhang, A. Chen, S. Wu, S. Yuan, and Y. Xiao. From AI Assistant to AI Scientist: Autonomous Discovery of LLM-RL Algorithms with LLM Agents, Mar. 2026. [find](https://scholar.google.com/scholar?q=S%2E+Xia%2C+Y%2E+Zhang%2C+A%2E+Chen%2C+S%2E+Wu%2C+S%2E+Yuan%2C+and+Y%2E+Xiao%2E+From+AI+Assistant+to+AI+Scientist%3A+Autonomous+Discovery+of+LLM%2DRL+Algorithms+with+LLM+Agents%2C+Mar%2E+2026%2E)
112. Z. Xie, Y. Guo, and K. Xu. AwesomeLit: Towards Hypothesis Generation with Agent-Supported Literature Research, Mar. 2026. [find](https://scholar.google.com/scholar?q=Z%2E+Xie%2C+Y%2E+Guo%2C+and+K%2E+Xu%2E+AwesomeLit%3A+Towards+Hypothesis+Generation+with+Agent%2DSupported+Literature+Research%2C+Mar%2E+2026%2E)
113. L. Xu and T. Borrett. Beyond ai as assistants: Toward autonomous discovery in cosmology. arXiv preprint arXiv:2605.14791, 2026. [find](https://scholar.google.com/scholar?q=L%2E+Xu+and+T%2E+Borrett%2E+Beyond+ai+as+assistants%3A+Toward+autonomous+discovery+in+cosmology%2E+arXiv+preprint+arXiv%3A2605%2E14791%2C+2026%2E)
114. L. Xu, Y. Zhou, and S. Pokutta. Agentic mip research: Accelerated constraint handler generation. arXiv preprint arXiv:2605.09186, 2026. — [arXiv:2605.09186](https://arxiv.org/abs/2605.09186) [find](https://scholar.google.com/scholar?q=L%2E+Xu%2C+Y%2E+Zhou%2C+and+S%2E+Pokutta%2E+Agentic+mip+research%3A+Accelerated+constraint+handler+generation%2E+arXiv+preprint+arXiv%3A2605%2E09186%2C+2026%2E+%E2%80%94+%5BarXiv%3A2605%2E09186%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2605%2E09186%29)
115. R. Xu and Y. Yan. Agent skills for large language models: Architecture, acquisition, security, and the path forward. arXiv preprint arXiv:2602.12430, 2026. — [arXiv:2602.12430](https://arxiv.org/abs/2602.12430) [find](https://scholar.google.com/scholar?q=R%2E+Xu+and+Y%2E+Yan%2E+Agent+skills+for+large+language+models%3A+Architecture%2C+acquisition%2C+security%2C+and+the+path+forward%2E+arXiv+preprint+arXiv%3A2602%2E12430%2C+2026%2E+%E2%80%94+%5BarXiv%3A2602%2E12430%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2)
116. W. Xu, T. Mi, Y. Liu, Y. Nan, Z. Zhou, L. Ye, L. Zhang, Y. Qiao, and P. Liu. ASI-Evolve: AI Accelerates AI, Mar. 2026. [find](https://scholar.google.com/scholar?q=W%2E+Xu%2C+T%2E+Mi%2C+Y%2E+Liu%2C+Y%2E+Nan%2C+Z%2E+Zhou%2C+L%2E+Ye%2C+L%2E+Zhang%2C+Y%2E+Qiao%2C+and+P%2E+Liu%2E+ASI%2DEvolve%3A+AI+Accelerates+AI%2C+Mar%2E+2026%2E)
117. Y. Yang, J. Li, Q. Pan, B. Zhan, Y. Cai, L. Du, J. Zhou, K. Chen, Q. Chen, X. Li, et al. Autoskill: Experience-driven lifelong learning via skill self-evolution. arXiv preprint arXiv:2603.01145, 2026. — [arXiv:2603.01145](https://arxiv.org/abs/2603.01145) [find](https://scholar.google.com/scholar?q=Y%2E+Yang%2C+J%2E+Li%2C+Q%2E+Pan%2C+B%2E+Zhan%2C+Y%2E+Cai%2C+L%2E+Du%2C+J%2E+Zhou%2C+K%2E+Chen%2C+Q%2E+Chen%2C+X%2E+Li%2C+et+al%2E+Autoskill%3A+Experience%2Ddriven+lifelong+learning+via+skill+self%2Devolution%2E+arXiv+preprint+arXiv%3A2603%2E01145%2C+2026%2E)
118. Z. Yang, X. Liu, and X. Xu. [Sciintegrity-bench: A benchmark for evaluating academic integrity in ai scientist systems](https://md.research.elamur.ai/free/en/papers/yang-2026-sciintegrity-bench-benchmark-evaluating/). arXiv preprint arXiv:2605.10246, 2026.
119. Z. Zhang, N. Wang, S. Galhotra, and C. Cardie. [How far are we from true auto-research?](https://md.research.elamur.ai/free/en/papers/zhang-2026-far-true-auto-research/) 2026.
120. H. Zhou, S. Guo, A. Liu, Z. Yu, Z. Gong, B. Zhao, Z. Chen, M. Zhang, Y. Chen, J. Li, et al. Memento-skills: Let agents design agents. arXiv preprint arXiv:2603.18743, 2026. — [arXiv:2603.18743](https://arxiv.org/abs/2603.18743) [find](https://scholar.google.com/scholar?q=H%2E+Zhou%2C+S%2E+Guo%2C+A%2E+Liu%2C+Z%2E+Yu%2C+Z%2E+Gong%2C+B%2E+Zhao%2C+Z%2E+Chen%2C+M%2E+Zhang%2C+Y%2E+Chen%2C+J%2E+Li%2C+et+al%2E+Memento%2Dskills%3A+Let+agents+design+agents%2E+arXiv+preprint+arXiv%3A2603%2E18743%2C+2026%2E+%E2%80%94+%5BarXiv%3A2603%2E18743%5D%28https%3A)
121. B. Zhu, Q. Jia, T. Lan, J. Ren, F. Gu, F. Jiang, L. Wang, Z. Xu, and W. Luo. Marco DeepResearch: Unlocking Efficient Deep Research Agents via Verification-Centric Design, Mar. 2026. [find](https://scholar.google.com/scholar?q=B%2E+Zhu%2C+Q%2E+Jia%2C+T%2E+Lan%2C+J%2E+Ren%2C+F%2E+Gu%2C+F%2E+Jiang%2C+L%2E+Wang%2C+Z%2E+Xu%2C+and+W%2E+Luo%2E+Marco+DeepResearch%3A+Unlocking+Efficient+Deep+Research+Agents+via+Verification%2DCentric+Design%2C+Mar%2E+2026%2E)
122. Q. Zou, H. H. Lam, W. Zhao, T. Chen, Y. Tang, S. Yu, Y. Zhu, S. Anumasa, Z. Zhang, T. Zhang, C. Liu, Z. Jiang, A. Goyal, and D. Liu. FML-bench: A controlled study of AI research agent strategies from the perspective of search dynamics. May 2026. [find](https://scholar.google.com/scholar?q=Q%2E+Zou%2C+H%2E+H%2E+Lam%2C+W%2E+Zhao%2C+T%2E+Chen%2C+Y%2E+Tang%2C+S%2E+Yu%2C+Y%2E+Zhu%2C+S%2E+Anumasa%2C+Z%2E+Zhang%2C+T%2E+Zhang%2C+C%2E+Liu%2C+Z%2E+Jiang%2C+A%2E+Goyal%2C+and+D%2E+Liu%2E+FML%2Dbench%3A+A+controlled+study+of+AI+research+agent+strategies+from+)

*Updated 3 November 2026*