Alien Science: Sampling Coherent but Cognitively Unavailable Research Directions from Idea Atoms
This addresses a specific creativity gap in AI research for generating novel research directions, though it is incremental as it builds on existing methods for idea decomposition and modeling.
The paper tackled the problem of generating research ideas that are coherent but non-obvious to the current community by formalizing cognitive availability and introducing a pipeline to sample such 'alien' directions, achieving more diverse outputs than LLM baselines while maintaining coherence on a corpus of ~7,500 recent LLM papers.
Large language models are adept at synthesizing and recombining familiar material, yet they often fail at a specific kind of creativity that matters most in research: producing ideas that are both coherent and non-obvious to the current community. We formalize this gap through cognitive availability, the likelihood that a research direction would be naturally proposed by a typical researcher given what they have worked on. We introduce a pipeline that (i) decomposes papers into granular conceptual units, (ii) clusters recurring units into a shared vocabulary of idea atoms, and (iii) learns two complementary models: a coherence model that scores whether a set of atoms constitutes a viable direction, and an availability model that scores how likely that direction is to be generated by researchers drawn from the community. We then sample "alien" directions that score high on coherence but low on availability. On a corpus of $\sim$7,500 recent LLM papers from NeurIPS, ICLR and ICML, we validate that (a) conceptual units preserve paper content under reconstruction, (b) idea atoms generalize across papers rather than memorizing paper-specific phrasing, and (c) the Alien sampler produces research directions that are more diverse than LLM baselines while maintaining coherence.