mlsanityData to 14 Oct 2026

Generating Literature-Driven Scientific Theories at Scale

arXiv 2601.16282 ↗First posted 22 Jan 2026cs.CL cs.AIcited by 4

Not on the map. Why?

Abstract (arXiv)

Contemporary automated scientific discovery has focused on agents for generating scientific experiments, while systems that perform higher-level scientific activities such as theory building remain underexplored. In this work, we formulate the problem of synthesizing theories consisting of qualitative and quantitative laws from large corpora of scientific literature. We study theory generation at scale, using 13.7k source papers to synthesize 2.9k theories, examining how generation using literature-grounding versus parametric knowledge, and accuracy-focused versus novelty-focused generation objectives change theory properties. Our experiments show that, compared to using parametric LLM memory for generation, our literature-supported method creates theories that are significantly better at both matching existing evidence and at predicting future results from 4.6k subsequently-written papers

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 22 of its 68 references that are in the map. First posted on arXiv 22 Jan 2026.

Builds on

22 of 68 references.

References · 22

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  2. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  3. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
  4. 2024The AI Scientist: Towards Fully Automated Open-Ended Scientific Discoverycited by 1,071
  5. 2019COMET: Commonsense Transformers for Automatic Knowledge Graph Constructioncited by 450
  6. 2025Agent Laboratory: Using LLM Agents as Research Assistantscited by 425
  7. 2024Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researcherscited by 363
  8. 2024ResearchAgent: Iterative Research Idea Generation over Scientific Literature with Large Language Modelscited by 210
  9. 2023SciMON: Scientific Inspiration Machines Optimized for Noveltycited by 179
  10. 2018Construction of the Literature Graph in Semantic Scholarcited by 118
  11. 2024BLADE: Benchmarking Language Model Agents for Data-Driven Sciencecited by 77
  12. 2019Inducing Relational Knowledge from BERTcited by 75
10 more
  1. 2025Robin: A multi-agent system for automating scientific discoverycited by 62
  2. 2025CodeScientist: End-to-End Semi-Automated Scientific Discovery with Code-based Experimentationcited by 57
  3. 2024Automated Statistical Model Discovery with Language Modelscited by 41
  4. 2025HypoBench: Towards Systematic and Principled Benchmarking for Hypothesis Generationcited by 37
  5. 2024Data-driven Discovery with Large Generative Modelscited by 34
  6. 2025AutoDiscovery: Open-ended Scientific Discovery via Bayesian Surprisecited by 33
  7. 2022Language Models as Inductive Reasonerscited by 32
  8. 2024Literature Meets Data: A Synergistic Approach to Hypothesis Generationcited by 20
  9. 2025HypER: Literature-grounded Hypothesis Generation and Distillation with Provenancecited by 10
  10. 2025HARPA: A Testability-Driven, Literature-Grounded Framework for Research Ideationcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026LLM-AutoSciLab: Closed-Loop Scientific Discovery via Active Experimentation with LLMscited by 12
  2. 2026Evidence-Informed LLM Beliefs for Continual Scientific Discoverycited by 1
  3. 2026Site4Drug: Predicting Drug-Binding Target Sites with an AI Agentcited by 0
  4. 2026Asterism: Exploring and Synthesizing Scattered Observations into Literature-Grounded Hypotheses and Theoriescited by 0