Words That Unite The World: A Unified Framework for Deciphering Central Bank Communications Globally

2025-05-15

arXiv.org

AI Summary

Central bank policy communication often carries ambiguous implicit meanings, and misinterpretations can exacerbate socioeconomic inequality. To address this, we propose the first globally harmonized analytical framework for central bank communication. We introduce the World Central Bank (WCB) dataset—comprising 380,000 sentences from 25 central banks over 28 years—and formalize three core tasks: stance detection, temporal classification, and uncertainty estimation. Methodologically, we pioneer a cross-central-bank joint modeling paradigm (“the whole exceeds the sum of its parts”) and design a high-fidelity semantic annotation protocol featuring dual-dimensional labeling and expert validation. Through systematic evaluation across seven pretrained language models and nine large language models (15,075 benchmark runs), we rigorously assess model generalizability and economic utility. All data, models, and code are publicly released under CC-BY-NC-SA 4.0 on Hugging Face and GitHub.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Central banks around the world play a crucial role in maintaining economic stability. Deciphering policy implications in their communications is essential, especially as misinterpretations can disproportionately impact vulnerable populations. To address this, we introduce the World Central Banks (WCB) dataset, the most comprehensive monetary policy corpus to date, comprising over 380k sentences from 25 central banks across diverse geographic regions, spanning 28 years of historical data. After uniformly sampling 1k sentences per bank (25k total) across all available years, we annotate and review each sentence using dual annotators, disagreement resolutions, and secondary expert reviews. We define three tasks: Stance Detection, Temporal Classification, and Uncertainty Estimation, with each sentence annotated for all three. We benchmark seven Pretrained Language Models (PLMs) and nine Large Language Models (LLMs) (Zero-Shot, Few-Shot, and with annotation guide) on these tasks, running 15,075 benchmarking experiments. We find that a model trained on aggregated data across banks significantly surpasses a model trained on an individual bank's data, confirming the principle"the whole is greater than the sum of its parts."Additionally, rigorous human evaluations, error analyses, and predictive tasks validate our framework's economic utility. Our artifacts are accessible through the HuggingFace and GitHub under the CC-BY-NC-SA 4.0 license.

Problem

• Deciphering policy implications in central bank communications globally
• Addressing misinterpretations that disproportionately impact vulnerable populations
• Benchmarking language models for stance detection and uncertainty estimation

Innovation

• Created comprehensive global central bank communications dataset
• Benchmarked pretrained and large language models extensively
• Demonstrated cross-bank training outperforms individual bank models