监测全球援助流量：一种使用大型语言模型的新方法（英）

Policy Research Working Paper11248Monitoring Global Aid FlowsA Novel Approach Using Large Language ModelsXubei LuoArvind Balaji RajasekaranAndrew Conner Scruggs Development Finance Vice Presidency November 2025 Public Disclosure AuthorizedPublic Disclosure AuthorizedPublic Disclosure AuthorizedPublic Disclosure AuthorizedProduced by the Research Support TeamAbstractThe Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.Policy Research Working Paper 11248Effective monitoring of development aid is the foundation for assessing the alignment of flows with their intended development objectives. Existing reporting systems, such as the Organisation for Economic Co-operation and Devel-opment’s Creditor Reporting System, provide standardized classification of aid activities but have limitations when it comes to capturing new areas like climate change, digitali-zation, and other cross-cutting themes. This paper proposes a bottom-up, unsupervised machine learning framework that leverages textual descriptions of aid projects to generate highly granular activity clusters. Using the 2021 Credi-tor Reporting System data set of nearly 400,000 records, the model produces 841 clusters, which are then grouped into 80 subsectors. These clusters reveal 36 emerging aid areas not tracked in the current Creditor Reporting System taxonomy, allow unpacking of “multi-sectoral” and “sector not specified” classifications, and enable estimation of flows to new themes, including World Bank Global Challenge Programs, International Development Association–20 Special Themes, and Cross-Cutting Issues. Validation against both Creditor Reporting System benchmarks and International Development Association commitment data demonstrates robustness. This approach illustrates how machine learning and the new advances in large language models can enhance the monitoring of global aid flows and inform future improvements in aid classification and reporting. It offers a useful tool that can support more responsive and evidence-based decision-making, helping to better align resources with evolving development priorities.This paper is a product of the Development Finance Vice Presidency. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Pape