mlsanityData to 7 Oct 2026

CausalEvolve: Towards Open-Ended Discovery with Causal Scratchpad

arXiv 2603.14575 ↗First posted 15 Mar 2026cs.LG cs.CL stat.MLcited by 3

Not on the map. Why?

Abstract (arXiv)

Evolve-based agent such as AlphaEvolve is one of the notable successes in using Large Language Models (LLMs) to build AI Scientists. These agents tackle open-ended scientific problems by iteratively improving and evolving programs, leveraging the prior knowledge and reasoning capabilities of LLMs. Despite the success, existing evolve-based agents lack targeted guidance for evolution and effective mechanisms for organizing and utilizing knowledge acquired from past evolutionary experience. Consequently, they suffer from decreasing evolution efficiency and exhibit oscillatory behavior when approaching known performance boundaries. To mitigate the gap, we develop CausalEvolve, equipped with a causal scratchpad that leverages LLMs to identify and reason about guiding factors for evolution. At the beginning, CausalEvolve first identifies outcome-level factors that offer complementary inspirations in improving the target objective. During the evolution, CausalEvolve also inspects surprise patterns during the evolution and abductive reasoning to hypothesize new factors, which in turn offer novel directions. Through comprehensive experiments, we show that CausalEvolve effectively improves the evolutionary efficiency and discovers better solutions in 4 challenging open-ended scientific tasks.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 46 of its 79 references that are in the map. First posted on arXiv 15 Mar 2026.

Builds on

46 of 79 references.

References · 46

  1. 2024The AI Scientist: Towards Fully Automated Open-Ended Scientific Discoverycited by 1,002
  2. 2025AlphaEvolve: A coding agent for scientific and algorithmic discoverycited by 763
  3. 2024MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineeringcited by 429
  4. 2025The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Searchcited by 375
  5. 2025Accelerating scientific discovery with Co-Scientistcited by 353
  6. 2025From System 1 to System 2: A Survey of Reasoning Large Language Modelscited by 319
  7. 2020Differentiable Causal Discovery from Interventional Datacited by 218
  8. 2023Can large language models provide useful feedback on research papers? A large-scale empirical analysiscited by 195
  9. 2023SciMON: Scientific Inspiration Machines Optimized for Noveltycited by 174
  10. 2019Causal Discovery from Heterogeneous/Nonstationary Data with Independent Changescited by 166
  11. 2013Causal Inference in the Presence of Latent Variables and Selection Biascited by 165
  12. 2025ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolutioncited by 146
34 more
  1. 2025Deep Research Agents: A Systematic Examination And Roadmapcited by 122
  2. 2025Agentic Large Language Models, a surveycited by 105
  3. 2020ICE-BeeM: Identifiable Conditional Energy-Based Deep Models Based on Nonlinear ICAcited by 102
  4. 2023Large Language Models for Automated Open-domain Scientific Hypotheses Discoverycited by 100
  5. 2024Chain of Ideas: Revolutionizing Research Via Novel Idea Development with LLM Agentscited by 94
  6. 2025From Automation to Autonomy: A Survey on Large Language Models in Scientific Discoverycited by 86
  7. 2025Kosmos: An AI Scientist for Autonomous Discoverycited by 78
  8. 2018Characterizing and Learning Equivalence Classes of Causal DAGs under Interventionscited by 76
  9. 2023Data-Copilot: Bridging Billions of Data and Humans with Autonomous Workflowcited by 76
  10. 2024MOOSE-Chem: Large Language Models for Rediscovering Unseen Chemistry Scientific Hypothesescited by 68
  11. 2025Mathematical exploration and discovery at scalecited by 66
  12. 2023Causal Discovery with Language Models as Imperfect Expertscited by 62
  13. 2022Causal Discovery in Heterogeneous Environments Under the Sparse Mechanism Shift Hypothesiscited by 54
  14. 2023SheetCopilot: Bringing Software Productivity to the Next Level through Large Language Modelscited by 53
  15. 2025ThetaEvolve: Test-time Learning on Open Problemscited by 52
  16. 2024Efficient Causal Graph Discovery Using Large Language Modelscited by 51
  17. 2025Barbarians at the Gate: How AI is Upending Systems Researchcited by 48
  18. 2019Causal Discovery and Forecasting in Nonstationary Environments with State-Space Modelscited by 45
  19. 2025Earth-Agent: Unlocking the Full Landscape of Earth Observation with Agentscited by 42
  20. 2025Automated Hypothesis Validation with Agentic Sequential Falsificationscited by 40
  21. 2023Understanding and Improving Feature Learning for Out-of-Distribution Generalizationcited by 35
  22. 2025Early science acceleration experiments with GPT-5cited by 35
  23. 2024ALCM: Autonomous LLM-Augmented Causal Discovery Frameworkcited by 29
  24. 2025DeepResearch Arena: The First Exam of LLMs' Research Abilities via Seminar-Grounded Taskscited by 24
  25. 2026Accelerating Scientific Research with Gemini: Case Studies and Common Techniquescited by 24
  26. 2023Causal Structure Learning Supervised by Large Language Modelcited by 21
  27. 2024End-To-End Causal Effect Estimation from Unstructured Natural Language Datacited by 18
  28. 2025Causal-Copilot: An Autonomous Causal Analysis Agentcited by 14
  29. 2024RealTCD: Temporal Causal Discovery from Interventional Data with Large Language Modelcited by 13
  30. 2024Discovering and Reasoning of Causality in the Hidden World with Large Language Modelscited by 10
  31. 2025Can Large Language Models Help Experimental Design for Causal Discovery?cited by 10
  32. 2025From What to Why: A Multi-Agent System for Evidence-based Chemical Reaction Condition Reasoningcited by 10
  33. 2024CriticAL: Critic Automation with Language Modelscited by 5
  34. 2025Revealing Multimodal Causality with Large Language Modelscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026What Do Evolutionary Coding Agents Evolve?cited by 3
  2. 2026Workflow Closure Is Not Scientific Closure in Auto-Research Systemscited by 1
  3. 2026LabBook: Harnessing Experimental History for Efficient LLM-Driven Discoverycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026InferenceEvolve: Towards Automated Causal Effect Estimators through Self-Evolving AI
  2. 2026EvoSCM: Scientific Belief Revision Through Causal Model Evolution and Experimentation
  3. 2026FreeEvolve: Learning to Evolve Beyond Fixed Loops
  4. 2026CausalGame: Benchmarking Causal Thinking of LLM Agents in Games
  5. 2025CodeEvolve: an open source evolutionary coding agent for algorithmic discovery and optimization
  6. 2025Evolution without an Oracle: Driving Effective Evolution with LLM Judges
  7. 2025Causal MAS: A Survey of Large Language Model Architectures for Discovery and Effect Estimation
  8. 2026AdaptEvolve: Improving Efficiency of Evolutionary AI Agents through Adaptive Model Selection