mlsanityData to 14 Oct 2026

Towards Scientific Discovery with Generative AI: Progress, Opportunities, and Challenges

arXiv 2412.11427 ↗First posted 16 Dec 2024cs.LG cs.AIcited by 41

Not on the map. Why?

Abstract (arXiv)

Scientific discovery is a complex cognitive process that has driven human knowledge and technological progress for centuries. While artificial intelligence (AI) has made significant advances in automating aspects of scientific reasoning, simulation, and experimentation, we still lack integrated AI systems capable of performing autonomous long-term scientific research and discovery. This paper examines the current state of AI for scientific discovery, highlighting recent progress in large language models and other AI techniques applied to scientific tasks. We then outline key challenges and promising research directions toward developing more comprehensive AI systems for scientific discovery, including the need for science-focused AI agents, improved benchmarks and evaluation metrics, multimodal scientific representations, and unified frameworks combining reasoning, theorem proving, and data-driven modeling. Addressing these challenges could lead to transformative AI tools to accelerate progress across disciplines towards scientific discovery.

Key facts. Cited by 41 papers in the map (data through 14 Oct 2026). Builds on 38 of its 67 references that are in the map. First posted on arXiv 16 Dec 2024.

Builds on

38 of 67 references.

References · 38

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  3. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
  4. 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,592
  5. 2019BioBERT: a pre-trained biomedical language representation model for biomedical text miningcited by 1,947
  6. 2022Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answeringcited by 1,919
  7. 2022Large Language Models Encode Clinical Knowledgecited by 1,830
  8. 2019SciBERT: A Pretrained Language Model for Scientific Textcited by 1,414
  9. 2024The AI Scientist: Towards Fully Automated Open-Ended Scientific Discoverycited by 1,071
  10. 2020Domain-Specific Language Model Pretraining for Biomedical Natural Language Processingcited by 1,009
  11. 2023ChemCrow: Augmenting large-language models with chemistry toolscited by 631
  12. 2019AI Feynman: a Physics-Inspired Method for Symbolic Regressioncited by 548
26 more
  1. 2022BioGPT: Generative Pre-trained Transformer for Biomedical Text Generation and Miningcited by 538
  2. 2023LeanDojo: Theorem Proving with Retrieval-Augmented Language Modelscited by 427
  3. 2020Lagrangian Neural Networkscited by 398
  4. 2020Generative Language Modeling for Automated Theorem Provingcited by 388
  5. 2024Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researcherscited by 363
  6. 2020Discovering Symbolic Models from Deep Learning with Inductive Biasescited by 344
  7. 2022Draft, Sketch, and Prove: Guiding Formal Theorem Provers with Informal Proofscited by 291
  8. 2023Reasoning or Reciting? Exploring the Capabilities and Limitations of Language Models Through Counterfactual Taskscited by 246
  9. 2022End-to-end symbolic regression with transformerscited by 203
  10. 2021Neural Symbolic Regression that Scalescited by 192
  11. 2017Learning to Plan Chemical Synthesescited by 179
  12. 2023SciMON: Scientific Inspiration Machines Optimized for Noveltycited by 179
  13. 2020AI Feynman 2.0: Pareto-optimal symbolic regression exploiting graph modularitycited by 144
  14. 2021Proof Artifact Co-training for Theorem Proving with Language Modelscited by 132
  15. 2022Semantic Probabilistic Layers for Neuro-Symbolic Learningcited by 103
  16. 2024SciAgents: Automating scientific discovery through multi-agent intelligent graph reasoningcited by 94
  17. 2023GIT-Mol: A Multi-modal Large Language Model for Molecular Science with Graph, Image, and Textcited by 82
  18. 2023Transformer-based Planning for Symbolic Regressioncited by 79
  19. 2023ProtST: Multi-Modality Learning of Protein Sequences and Biomedical Textscited by 78
  20. 2024CRISPR-GPT for Agentic Automation of Gene-editing Experimentscited by 58
  21. 2023SNIP: Bridging Mathematical Symbolic and Numeric Realms with Unified Pre-trainingcited by 47
  22. 2024Are LLMs Ready for Real-World Materials Discovery?cited by 40
  23. 2024AtomAgents: Alloy design and discovery through physics-aware multi-modal multi-agent artificial intelligencecited by 36
  24. 2024Meta-Designing Quantum Experiments with Language Modelscited by 12
  25. 2024Leveraging Chemistry Foundation Models to Facilitate Structure Focused Retrieval Augmented Generation in Multi-Agent Workflows for Catalyst and Materials Designcited by 3
  26. 2024Knowledge-Reuse Transfer Learning Methods in Molecular and Material Sciencecited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 23; 2026: 18.

41 papers.

Sort:

By area:

  1. 2025AI-Researcher: Autonomous Scientific Innovationcited by 100
  2. 2025LLM-SRBench: A New Benchmark for Scientific Equation Discovery with Large Language Modelscited by 71
  3. 2026EvoScientist: Towards Multi-Agent Evolving AI Scientists for End-to-End Scientific Discoverycited by 51
  4. 2025Position: Multimodal Large Language Models Can Significantly Advance Scientific Reasoningcited by 33
  5. 2025PiFlow: Principle-Aware Scientific Discovery with Multi-Agent Collaborationcited by 28
  6. 2025PhysGym: Benchmarking LLMs in Interactive Physics Discovery with Controlled Priorscited by 15
  7. 2025Are Large Language Models Reliable AI Scientists? Assessing Reverse-Engineering of Black-Box Systemscited by 14
  8. 2025LLEMA: Evolutionary Search with LLMs for Multi-Objective Materials Discoverycited by 13
  9. 2025Agentic Deep Graph Reasoning Yields Self-Organizing Knowledge Networkscited by 12
  10. 2026LLM-AutoSciLab: Closed-Loop Scientific Discovery via Active Experimentation with LLMscited by 12
  11. 2025Autonomous Agents for Scientific Discovery: Orchestrating Scientists, Language, Code, and Physicscited by 11
  12. 2025AtmosSci-Bench: Evaluating the Recent Advance of Large Language Model for Atmospheric Sciencecited by 9
29 more
  1. 2025Benchmarking AI scientists for omics data driven biological discoverycited by 7
  2. 2025Mimicking the Physicist's Eye:A VLM-centric Approach for Physics Formula Discoverycited by 7
  3. 2025The Need for Verification in AI-Driven Scientific Discoverycited by 7
  4. 2025Deep Ideation: Designing LLM Agents to Generate Novel Research Ideas on Scientific Concept Networkcited by 7
  5. 2026Towards a Medical AI Scientistcited by 7
  6. 2026LLM-ACES: Closed-Loop Discovery of Dynamical Systems with LLM-Guided Adaptive Searchcited by 7
  7. 2025Diffusion Models for Future Networks and Communications: A Comprehensive Surveycited by 6
  8. 2025ATLAS: A High-Difficulty, Multidisciplinary Benchmark for Frontier Scientific Reasoningcited by 6
  9. 2025AI-Newton: A Concept-Driven Physical Law Discovery System without Prior Physical Knowledgecited by 5
  10. 2026Geometry-Calibrated Conformal Abstention for Language Modelscited by 5
  11. 2025AI-Driven Automation Can Become the Foundation of Next-Era Science of Science Researchcited by 3
  12. 2026ProjectionBench: Evaluating Scientific Hypothesis Generation in LLMs Under Progressive Information Disclosurecited by 3
  13. 2026Position: Academic Conferences are Potentially Facing Denominator Gaming Caused by Fully Automated Scientific Agentscited by 2
  14. 2025SURFACEBENCH: A Geometry-Aware Benchmark for Symbolic Surface Discoverycited by 1
  15. 2025From Verification Burden to Trusted Collaboration: Design Goals for LLM-Assisted Literature Reviewscited by 1
  16. 2026SciHorizon-GENE: Benchmarking LLM for Life Sciences Inference from Gene Knowledge to Functional Understandingcited by 1
  17. 2026Beyond the Desk: Barriers and Future Opportunities for AI to Assist Scientists in Embodied Physical Taskscited by 1
  18. 2026AwesomeLit: Towards Hypothesis Generation with Agent-Supported Literature Researchcited by 1
  19. 2026SciPaths: Forecasting Pathways to Scientific Discoverycited by 1
  20. 2026Can AI Follow In Einstein's Footsteps?cited by 1
  21. 2026The Greatness of Science Cannot Be Planned: Agentic Auto-Research is Fuzz Testingcited by 1
  22. 2025Bridging the Gap Between Scientific Laws Derived by AI Systems and Canonical Knowledge via Abductive Inference with AI-Noethercited by 0
  23. 2025AI Driven Discovery of Bio Ecological Mediation in Cascading Heatwave Riskscited by 0
  24. 2025Project Rachel: Can an AI Become a Scholarly Author?cited by 0
  25. 2026More Than Can Be Said: A Benchmark and Framework for Pre-Question Scientific Ideationcited by 0
  26. 2026TopoAlign: Topology-Aware Visual Representation Alignmentcited by 0
  27. 2026Smaller, Younger, and More Impactful: How AI-Assisted Writing Transforms Research Teamscited by 0
  28. 2026IDEAFix: Evaluation Framework for Creative Defixation Prompting in LLMscited by 0
  29. 2026EvoMO-SR: Multiobjective LLM-based Evolution of Symbolic Expressions with substructure guidancecited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Transforming Science with Large Language Models: A Survey on AI-assisted Scientific Discovery, Experimentation, Content Generation, and Evaluation
  2. 2025From AI for Science to Agentic Science: A Survey on Autonomous Scientific Discovery
  3. 2025Agentic AI for Scientific Discovery: A Survey of Progress, Challenges, and Future Directions
  4. 2025Virtuous Machines: Towards Artificial General Science
  5. 2025The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search
  6. 2024DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agents
  7. 2026Contemporary AI lacks the imagination to diverge or negate in science
  8. 2025AI4Research: A Survey of Artificial Intelligence for Scientific Research