Abstract:Scientific contributions rarely develop in isolation, but instead build upon prior discoveries. We formulate the task of automated technological roadmapping as extracting scientific contributions from scholarly articles and linking them to their prerequisites. We present the Scientific Contribution Graph, a large-scale AI/NLP-domain resource containing 2 million detailed scientific contributions extracted from 230k open-access papers and connected by 12.5 million prerequisite edges. We further introduce scientific prerequisite prediction, a scientific discovery task in which models predict which existing technologies can enable future discoveries, and show that contemporary models are rapidly improving on this task, reaching 0.48 MAP when evaluated using temporally filtered backtesting. We anticipate technological roadmapping resources such as this will support scientific impact assessment and automated scientific discovery.
| Comments: | 8 pages, 5 figures |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2605.15011 [cs.CL] |
| (or arXiv:2605.15011v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.15011 arXiv-issued DOI via DataCite |
From: Peter Jansen [view email]
[v1]
Thu, 14 May 2026 16:12:12 UTC (515 KB)
[v2]
Sat, 23 May 2026 23:25:12 UTC (589 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。