FAME: Forecasting Academic Impact via Continuous-Time Manifold Evolution
About
Large Language Models (LLMs) are increasingly used to brainstorm and evaluate research ideas, yet assessing such judgments is fundamentally difficult because the true impact of a new idea may take years to emerge. We address this challenge by using the impact forecasting of human-authored manuscripts as a verifiable proxy task. In a prospective forecasting study, we find that frontier LLMs fail to reliably distinguish high-impact papers from ordinary publications, suggesting that static text-based judging is insufficient for scientific evaluation. To address this limitation, we propose $\textbf{FAME}$ ($\underline{\text{F}}$orecasting $\underline{\text{A}}$cademic Impact via Continuous-Time $\underline{\text{M}}$anifold $\underline{\text{E}}$volution), a spatiotemporal framework for modeling the dynamic trajectories of scientific topics. FAME projects papers into a dynamic latent space informed by textual features and a verified knowledge-flow graph, learning geometric constraints that align impactful manuscripts with the forward momentum of their fields. Experiments on 3,200 arXiv papers across three fast-evolving subfields show that FAME consistently and substantially outperforms state-of-the-art LLM evaluators in prospective multidimensional impact forecasting. Furthermore, integrating FAME's dynamic geometric signals into LLMs significantly improves their forecasting performance. These results support manuscript impact forecasting as a useful, measurable proxy benchmark and position FAME as a strong, trajectory-aware foundation for automated scientific evaluation.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Prospective paper impact forecasting | arXiv August 2025 | Top-5 Accuracy60 | 12 | |
| Prospective paper impact forecasting | arXiv September 2025 | Top-5 Accuracy60 | 12 | |
| Prospective paper impact forecasting | arXiv October 2025 | Top-5 Accuracy46.7 | 12 | |
| Prospective paper impact forecasting | arXiv (November 2025) | Top-5 Accuracy33.3 | 12 | |
| Prospective paper impact forecasting | arXiv June 2025 - November 2025 Average | Top-5 Accuracy58.9 | 12 | |
| Scientific Impact Forecasting | arXiv June 2024 to November 2025 (temporal out-of-distribution (OOD)) | Forecast Score (2025.08)0.582 | 12 | |
| Prospective paper impact forecasting | arXiv January 2025 | Top-5 Accuracy66.7 | 11 | |
| Prospective paper impact forecasting | arXiv February 2025 | Top-5 Accuracy80 | 11 | |
| Prospective paper impact forecasting | arXiv (March 2025) | Top-5 Accuracy60 | 11 | |
| Prospective paper impact forecasting | arXiv April 2025 | Top-5 Accuracy66.7 | 11 |