Can Large Language Models Review the Scientific Literature? A Test Against a 102-Article Benchmark
arXiv:2611.05459v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to support scientific literature reviews, yet evaluations often examine individual tasks rather than the review process as a whole. We assess six AI tools across four stages: finding relevant articles, interpreting them, synthesizing across them, and proposing future research. The benchmark is a published review of the Spanish plume weather pattern and its 102-article corpus. In the initial searche
阅读原始来源
来源记录
研究分析
研究主张
论文标识：2611.05459。 来源描述：arXiv:2611.05459v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to support scientific literature reviews, yet evaluations often examine individual tasks rather than the review process as a whole. We assess six AI tools across four stages: finding relevant articles, interpreting them, synthesizing across them, and proposing future research. The benchmark is a published review of the Spanish plume weather pattern and its 102-article corpus. In the initial searche
评测与比较记录
可供比较的近期同类记录：“PharmAgents2: Towards Autonomous End-to-End Drug Discovery” (arXiv, 2026年11月6日)；“Align Before Aggregation: Basis-Consistent Federated LoRA under Heterogeneous Ranks” (arXiv, 2026年11月6日)。主题相近仅用于导航，未核验其是否属于同一需求或独立证据。
结论适用范围
结论只能在所述模型版本、任务、数据与推理预算内解释。社区票数、下载量或榜单位置不能证明另一工作流中的可靠性。
有效性与复现缺口
需核对样本选择、基线配置、不确定性、数据污染风险、负面结果以及代码或权重可获取性；此处没有本站复现实验记录。
复核路线
阅读论文或模型卡及评测附录，针对明确基线复现一个有限结论，再测试代表性用户任务并记录结论失效的场景。