arXiv cs.CL· Yixuan Tang, Yi Yang·· 2 天前AI 评分22
用大语言模型从 FOMC 声明中解码货币政策立场：Delta-Consistent Scoring 框架
Mind the Shift: Decoding Monetary Policy Stance from FOMC Statements with Large Language Models
AI 导读
研究者提出 Delta-Consistent Scoring（DCS），一种无需人工标注的框架，将冻结的大语言模型表示映射为连续立场分数，同时建模声明的绝对立场与相邻会议间的相对变化。在四个 LLM 主干上，DCS 持续优于监督探针和 LLM-as-judge 基线，句级鹰派—鸽派分类准确率最高达 71.1%。其会议级分数与通胀指标高度相关，并与美国国债收益率变动显著关联。
正文 · 原文
Abstract:Federal Open Market Committee (FOMC) statements are a major source of monetary-policy information, and even subtle changes in their wording can move global financial markets. A central task is therefore to measure the hawkish--dovish stance conveyed in these texts. Existing approaches typically treat stance detection as a standard classification problem, labeling each statement in isolation. However, the interpretation of monetary-policy communication is inherently relative: market reactions depend not only on the tone of a statement, but also on how that tone shifts across meetings. We introduce Delta-Consistent Scoring (DCS), an annotation-free framework that maps frozen large language model (LLM) representations to continuous stance scores by jointly modeling absolute stance and relative inter-meeting shifts. Rather than relying on manual hawkish--dovish labels, DCS uses consecutive meetings as a source of self-supervision. It learns an absolute stance score for each statement and a relative shift score between consecutive statements. A delta-consistency objective encourages changes in absolute scores to align with the relative shifts. This allows DCS to recover a temporally coherent stance trajectory without manual labels. Across four LLM backbones, DCS consistently outperforms supervised probes and LLM-as-judge baselines, achieving up to 71.1% accuracy on sentence-level hawkish--dovish classification. The resulting meeting-level scores are also economically meaningful: they correlate strongly with inflation indicators and are significantly associated with Treasury yield movements. Overall, the results suggest that LLM representations encode monetary-policy signals that can be recovered through relative temporal structure.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2603.14313 [cs.CL] |
| (or arXiv:2603.14313v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2603.14313 arXiv-issued DOI via DataCite |
Submission history
From: Yixuan Tang [view email]
[v1]
Sun, 15 Mar 2026 10:07:28 UTC (58 KB)
[v2]
Thu, 15 Oct 2026 08:57:59 UTC (544 KB)
来源：arXiv cs.CL · arxiv.org