Hacker News 熱門（buzzing.cc 中文翻譯）·· 4 小時前AI 評分58
科學中的「異質空間」：連貫但認知上不可及的方向
科学中的“异质空间”：连贯但认知上不可及的方向
AI 導讀
研究提出探索科學「異質空間」的方法，將論文拆解為概念單元，評估概念組合的研究連貫性，以及既有研究社羣提出該方向的可能性。在16,068篇來自NeurIPS、ICLR、ICML及主要NLP會議的經同行評審LLM論文上，方法產生的研究組合比前沿LLM基線（包括可查看整批輸出的批次提示版本）更多元；盲評的人類與LLM評估及下游實驗評估顯示，每項構想的質素相若或更高。
正文
Abstract:Scientific discovery is constrained not only by what is true, but by what is cognitively available to researchers. Many directions are coherent in light of the literature yet unlikely to be proposed because no existing community combines the relevant concepts, methods, and intuitions. Modern language models inherit this bias, recombining high-density regions of the literature when prompted for novel ideas. We target the complementary region, the alien space of science: directions that are scientifically plausible but unlikely under the distribution of existing researchers. Our method decomposes papers into conceptual units and clusters them into a vocabulary of idea atoms. A coherence model scores whether atom combinations form viable research directions, while an availability model scores whether any existing author community is positioned to produce them. We rank combinations to maximize coherence while minimizing availability. On 16,068 peer-reviewed LLM papers from NeurIPS, ICLR, ICML, and major NLP venues, the sampler produces more diverse research portfolios than frontier LLM baselines, including batch-prompted variants that see their entire output, while matching or exceeding their per-idea quality under blind human, LLM, and downstream experimental evaluation. In temporal validation, all components use only pre-2025 data. Selecting less than 1% of the candidate space recovers 2025 combinations at 35.9 times chance; weighting unavailability increases recovery of strictly novel combinations to 3.24 times chance. By separating scientific plausibility from community availability, our framework points toward AI ideation that complements rather than merely accelerates human science, expanding exploration into coherent directions the community may overlook.
| Comments: | Advances in Neural Information Processing Systems, NeurIPS 2026 Main Track. 11 main pages, 44 appendix pages, 29 figures |
| Subjects: | Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2603.01092 [cs.AI] |
| (or arXiv:2603.01092v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2603.01092 arXiv-issued DOI via DataCite |
Submission history
From: Alejandro Hernandez Artiles [view email]
[v1]
Sun, 1 Mar 2026 13:05:19 UTC (569 KB)
[v2]
Fri, 15 May 2026 23:12:34 UTC (2,833 KB)
[v3]
Wed, 21 Oct 2026 08:59:34 UTC (3,197 KB)
來源：Hacker News 熱門（buzzing.cc 中文翻譯） · arxiv.org