FinCoT: Grounding Chain-of-Thought in Expert Financial Reasoning

TL;DR

FinCoT introduces a structured, expert-guided chain-of-thought prompting framework for financial NLP, significantly enhancing model accuracy, interpretability, and efficiency across multiple financial reasoning tasks.

Contribution

This work is the first to incorporate domain-specific expert blueprints into structured chain-of-thought prompting for financial NLP, improving performance and interpretability.

Findings

FinCoT increases model accuracy from 63.2% to 80.5%.

FinCoT reduces output length by up to 8.9x.

FinCoT enhances interpretability and aligns reasoning with domain expertise.

Abstract

This paper presents FinCoT, a structured chain-of-thought (CoT) prompting framework that embeds domain-specific expert financial reasoning blueprints to guide large language models' behaviors. We identify three main prompting styles in financial NLP (FinNLP): (1) standard prompting (zero-shot), (2) unstructured CoT (free-form reasoning), and (3) structured CoT (with explicitly structured reasoning steps). Prior work has mainly focused on the first two, while structured CoT remains underexplored and lacks domain expertise incorporation. Therefore, we evaluate all three prompting approaches across ten CFA-style financial domains and introduce FinCoT as the first structured finance-specific prompting approach incorporating blueprints from domain experts. FinCoT improves the accuracy of a general-purpose model, Qwen3-8B-Base, from 63.2% to 80.5%, and boosts Fin-R1 (7B), a finance-specific model, from 65.7% to 75.7%, while reducing output length by up to 8.9x and 1.16x compared to structured CoT methods, respectively. We find that FinCoT proves most effective for models lacking financial post-training. Our findings show that FinCoT does not only improve performance and reduce inference costs but also yields more interpretable and expert-aligned reasoning traces.