The Alien Space of Science: Sampling Coherent but Cognitively Unavailable Research Directions

A way to search for scientifically coherent ideas that existing research communities are unlikely to propose.

PreprintScientific ideationRevised May 15, 2026

Read on arXivCode and data

When we ask a language model for research ideas, how often does it return to the same familiar combinations? We wanted to search beyond the directions that a field is already equipped to imagine.

Alien Science scores two things separately: whether an idea is scientifically coherent, and whether an existing research community is likely to propose it. We search for ideas that score well on the first and low on the second.

How it works

The method decomposes papers into reusable idea atoms, learns whether combinations of those atoms form coherent research directions, and separately estimates whether an existing author community is positioned to produce them. It then ranks combinations that are coherent but comparatively unavailable.

Atomize

Decompose papers into a shared vocabulary of reusable scientific ideas.

Score coherence

Estimate whether a combination could form a viable research direction.

Score availability

Estimate whether an existing author community is positioned to propose it.

Sample

Search for directions that remain coherent while being comparatively unavailable.

Schematic of the search target: coherent ideas that existing research communities are unlikely to propose.

What to keep in mind

Community availability is inferred from publication and authorship patterns, which are imperfect proxies for what researchers know or could imagine.

The reported study focuses on the LLM literature; extending the method to other scientific fields requires new vocabularies and evaluation.

Two questions for the same idea

Could this combination of ideas form a plausible study? Is an existing author community positioned to propose it? The method learns a separate score for each question.

See the sampling method

Results at a glance

3.5–7×Broader effective atom vocabularyCompared with frontier-LLM ideation baselines
16,068Peer-reviewed LLM papersAcross major ML and NLP venues
3Blind evaluation modesLLM, human, and downstream experiments

The study covers LLM research. A broader idea vocabulary does not establish that every proposed direction is novel or feasible; domain experts still need to assess the ideas.

Source: Alien Science v2, evaluation

People behind the work

Listed in the paper’s author order.

Alejandro H. Artiles

Tiptree Systems · Max Planck Institute for Human Development · Max Planck Institute for Intelligent Systems

First author of The Alien Space of Science. Now studies cognition–action tradeoffs in language agents through TextCraft.

Contributed to

LacunaThe Alien Space of ScienceTextCraft

Martin Weiss

Tiptree Systems · Mila · Polytechnique Montréal

Builds tools that connect the literature, research questions, and the people working on them. Also studies peer-review verification and agent decision-making.

Contributed to

LacunaThe Alien Space of ScienceAI Meta-ReviewingLLM Micro-Rationality

Levin Brinkmann

Max Planck Institute for Human Development

Research collaborator at the Max Planck Institute for Human Development. Coauthor of The Alien Space of Science.

Contributed to

The Alien Space of Science

Iyad Rahwan

Max Planck Institute for Human Development

Coauthor of Alien Science at the Max Planck Institute for Human Development, studying how AI can expand scientific exploration.

Contributed to

The Alien Space of Science

Bernhard Schölkopf

Max Planck Institute for Intelligent Systems · ELLIS Institute Tübingen

Scientific advisor and coauthor of Alien Science, exploring research directions beyond familiar combinations of ideas.

Contributed to

The Alien Space of Science

Christopher Pal

Polytechnique Montréal · Mila · Canada CIFAR AI Chair

Canada CIFAR AI Chair at Polytechnique Montréal. Scientific advisor and coauthor of Lacuna and The Alien Space of Science.

Contributed to

LacunaThe Alien Space of Science

Hugo Larochelle

Scientific Director, Mila · Université de Montréal · McGill University

Scientific advisor and coauthor of Lacuna and Alien Science. Also collaborates on the behavioural economics of LLM agents.

Contributed to

LacunaThe Alien Space of ScienceLLM Micro-Rationality

Anirudh Goyal

Meta Superintelligence Labs · Formerly Google DeepMind

Research collaborator at Meta Superintelligence Labs and coauthor of The Alien Space of Science.

Contributed to

The Alien Space of Science

Nasim Rahaman

Tiptree Systems

Brings a background in theoretical physics and machine learning to scientific discovery. Coauthor of Lacuna and The Alien Space of Science.

Contributed to

LacunaThe Alien Space of Science

Read, reuse and cite

The paper and its artifacts, and a citation ready for your reference manager.

Cite this work

Copy this entry into your .bib file or import it into your reference manager.

@misc{artiles2026alienspace,
  title = {The Alien Space of Science: Sampling Coherent but Cognitively Unavailable Research Directions},
  author = {Artiles, Alejandro H. and Weiss, Martin and Brinkmann, Levin and Rahwan, Iyad and Sch{\"o}lkopf, Bernhard and Pal, Christopher and Larochelle, Hugo and Goyal, Anirudh and Rahaman, Nasim},
  year = {2026},
  publisher = {arXiv},
  doi = {10.48550/arXiv.2603.01092},
  url = {https://arxiv.org/abs/2603.01092}
}

Read the full paper

Open the PDF or preview it here.

Open PDF

Have a question worth working on?

We work with researchers on scientific discovery, agent evaluation, and tools for navigating the literature. Tell us what you are exploring.

Talk research with us