mlsanityData to 7 Oct 2026

The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search

arXiv 2504.08066 ↗First posted 10 Apr 2025cs.AI cs.CL cs.LGcited by 375

On the map: LLM agents and tool use → Web, search, and research agents

Abstract (arXiv)

AI is increasingly playing a pivotal role in transforming how scientific discoveries are made. We introduce The AI Scientist-v2, an end-to-end agentic system capable of producing the first entirely AI generated peer-review-accepted workshop paper. This system iteratively formulates scientific hypotheses, designs and executes experiments, analyzes and visualizes data, and autonomously authors scientific manuscripts. Compared to its predecessor (v1, Lu et al., 2024 arXiv:2408.06292), The AI Scientist-v2 eliminates the reliance on human-authored code templates, generalizes effectively across diverse machine learning domains, and leverages a novel progressive agentic tree-search methodology managed by a dedicated experiment manager agent. Additionally, we enhance the AI reviewer component by integrating a Vision-Language Model (VLM) feedback loop for iterative refinement of content and aesthetics of the figures. We evaluated The AI Scientist-v2 by submitting three fully autonomous manuscripts to a peer-reviewed ICLR workshop. Notably, one manuscript achieved high enough scores to exceed the average human acceptance threshold, marking the first instance of a fully AI-generated paper successfully navigating a peer review. This accomplishment highlights the growing capability of AI in conducting all aspects of scientific research. We anticipate that further advancements in autonomous scientific discovery technologies will profoundly impact human knowledge generation, enabling unprecedented scalability in research productivity and significantly accelerating scientific breakthroughs, greatly benefiting society at large. We have open-sourced the code at https://github.com/SakanaAI/AI-Scientist-v2 to foster the future development of this transformative technology. We also discuss the role of AI in science, including AI safety.

Key facts. Cited by 375 papers in the map (data through 7 Oct 2026). Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use. Builds on 10 of its 31 references that are in the map. First posted on arXiv 10 Apr 2025.

Builds on

10 of 31 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Agent Laboratory: Using LLM Agents as Research Assistants8 Jan

Other references · 9

  1. 2024The AI Scientist: Towards Fully Automated Open-Ended Scientific Discoverycited by 1,002
  2. 2025Accelerating scientific discovery with Co-Scientistcited by 353
  3. 2025AIDE: AI-Driven Exploration in the Space of Codecited by 206
  4. 2019AI-GAs: AI-generating algorithms, an alternate paradigm for producing general artificial intelligencecited by 84
  5. 2025BixBench: a Comprehensive Benchmark for LLM-based Agents in Computational Biologycited by 75
  6. 2025Superintelligent Agents Pose Catastrophic Risks: Can Scientist AI Offer a Safer Path?cited by 70
  7. 2025AgentRxiv: Towards Collaborative Autonomous Researchcited by 64
  8. 2025Transforming Science with Large Language Models: A Survey on AI-assisted Scientific Discovery, Experimentation, Content Generation, and Evaluationcited by 41
  9. 2025Evaluating Sakana's AI Scientist: Bold Claims, Mixed Results, and a Promising Future?cited by 32

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 88; 2026: 287.

375 papers.

Sort:

By area:

  1. 2025Perception, Reason, Think, and Plan: A Survey on Large Multimodal Reasoning Modelscited by 89
  2. 2025AI-Researcher: Autonomous Scientific Innovationcited by 88
  3. 2025From Automation to Autonomy: A Survey on Large Language Models in Scientific Discoverycited by 86
  4. 2026Self-Harness: Harnesses That Improve Themselvescited by 79
  5. 2025From AI for Science to Agentic Science: A Survey on Autonomous Scientific Discoverycited by 74
  6. 2025Paper2Code: Automating Code Generation from Scientific Papers in Machine Learningcited by 67
  7. 2025AI Research Agents for Machine Learning: Search, Exploration, and Generalization in MLE-benchcited by 65
  8. 2025Deep Research: A Survey of Autonomous Research Agentscited by 63
  9. 2025A Self-Improving Coding Agentcited by 58
  10. 2025TRiSM for Agentic AI: A Review of Trust, Risk, and Security Management in LLM-based Agentic Multi-Agent Systemscited by 58
  11. 2025MLR-Bench: Evaluating AI Agents on Open-Ended Machine Learning Researchcited by 57
  12. 2025DeepScientist: Advancing Frontier-Pushing Scientific Findings Progressivelycited by 56
38 more
  1. 2025Towards Robust Agentic CUDA Kernel Benchmarking, Verification, and Optimizationcited by 52
  2. 2026A Survey of Agentic Reasoning for Large Language Models: Towards Recursively Self-Improving and Collective Agentscited by 47
  3. 2026Code as Agent Harnesscited by 40
  4. 2025ReportBench: Evaluating Deep Research Agents via Academic Survey Taskscited by 34
  5. 2025A Survey of Scientific Large Language Models: From Data Foundations to Agent Frontierscited by 34
  6. 2025NewtonBench: Benchmarking Generalizable Scientific Law Discovery in LLM Agentscited by 32
  7. 2026AIRS-Bench: a Suite of Tasks for Frontier AI Research Science Agentscited by 32
  8. 2026ENPIRE: Agentic Robot Policy Self-Improvement in the Real Worldcited by 31
  9. 2025MLRC-Bench: Can Language Agents Solve Machine Learning Research Challenges?cited by 27
  10. 2025Ax-Prover: A Deep Reasoning Agentic Framework for Theorem Proving in Mathematics and Quantum Physicscited by 27
  11. 2026The 2025 AI Agent Index: Documenting Technical and Safety Features of Deployed Agentic AI Systemscited by 27
  12. 2025PiFlow: Principle-Aware Scientific Discovery with Multi-Agent Collaborationcited by 25
  13. 2025MM-Agent: LLM as Agents for Real-world Mathematical Modeling Problemcited by 24
  14. 2026InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discoverycited by 24
  15. 2026AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaborationcited by 23
  16. 2026MARS: Modular Agent with Reflective Search for Automated AI Researchcited by 22
  17. 2025AI Scientists Fail Without Strong Implementation Capabilitycited by 20
  18. 2025AutoMind: Adaptive Knowledgeable Agent for Automated Data Sciencecited by 20
  19. 2026AutoFigure: Generating and Refining Publication-Ready Scientific Illustrationscited by 20
  20. 2026Towards Execution-Grounded Automated AI Researchcited by 19
  21. 2026AutoSOTA: An End-to-End Automated Research System for State-of-the-Art AI Model Discoverycited by 17
  22. 2026Self-Improvements in Modern Agentic Systems: A Surveycited by 17
  23. 2025Accelerating Scientific Discovery with Autonomous Goal-evolving Agentscited by 16
  24. 2026Learning to Continually Learn via Meta-learning Agentic Memory Designscited by 16
  25. 2026The Red Queen G"odel Machine: Co-Evolving Agents and Their Evaluatorscited by 16
  26. 2025How Far Are AI Scientists from Changing the World?cited by 15
  27. 2025OmniScientist: Toward a Co-evolving Ecosystem of Human and AI Scientistscited by 15
  28. 2026ResearchGym: Evaluating Language Model Agents on Real-World AI Researchcited by 15
  29. 2026Toward Autonomous Long-Horizon Engineering for ML Researchcited by 15
  30. 2026ScientistOne: Towards Human-Level Autonomous Research via Chain-of-Evidencecited by 15
  31. 2025Predicting Empirical AI Research Outcomes with Language Modelscited by 14
  32. 2025PhysGym: Benchmarking LLMs in Interactive Physics Discovery with Controlled Priorscited by 13
  33. 2026FIRE-Bench: Evaluating AI Agents on the Rediscovery of Scientific Insightscited by 13
  34. 2026Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memorycited by 13
  35. 2026AI co-mathematician: Accelerating mathematicians with agentic AIcited by 13
  36. 2025The Automated LLM Speedrunning Benchmark: Reproducing NanoGPT Improvementscited by 12
  37. 2026Why LLMs Aren't Scientists Yet: Lessons from Four Autonomous Research Attemptscited by 12
  38. 2026The Last Human-Written Paper: Agent-Native Research Artifactscited by 12

50 most cited of 375 All 375 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Towards End-to-End Automation of AI Research
  2. 2025Agentic AI for Scientific Discovery: A Survey of Progress, Challenges, and Future Directions
  3. 2026Towards Automating Scientific Review with Google's Paper Assistant Tool
  4. 2026AIBuildAI-2: A Knowledge-Enhanced Agent for Automatically Building AI Models
  5. 2025Virtuous Machines: Towards Artificial General Science
  6. 2025aiXiv: A Next-Generation Open Access Ecosystem for Scientific Discovery Generated by AI Scientists
  7. 2026ResearchClawBench: A Benchmark for End-to-End Autonomous Scientific Research
  8. 2024ScienceAgentBench: Toward Rigorous Assessment of Language Agents for Data-Driven Scientific Discovery