CreditARF: A Framework for Corporate Credit Rating with Annual Report and Financial Feature Integration

2025-08-02

AI Summary

Existing corporate credit rating models predominantly rely on financial metrics and deep learning, overlooking rich credit signals embedded in unstructured textual sources such as annual reports. Method: This paper proposes the first multimodal credit rating framework integrating structured financial data and unstructured annual report text. It employs FinBERT to extract semantic features from textual content and introduces a novel dual-stream feature fusion mechanism that jointly models financial and textual modalities. Additionally, we construct CCRD—the first large-scale, publicly available, comprehensive dataset for multimodal credit rating. Contribution/Results: Experiments demonstrate consistent performance gains across multiple benchmarks, improving rating accuracy by 8–12%. The proposed framework significantly enhances model discriminability and robustness. This work provides the first systematic empirical validation of the predictive value of non-financial textual information in credit assessment, establishing a reproducible methodology and foundational dataset for multi-source, heterogeneous-data-driven intelligent risk control.

Technology Category

• Machine Learning: Multimodal Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Computer Vision: Multi-modal Vision

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

Corporate credit rating serves as a crucial intermediary service in the market economy, playing a key role in maintaining economic order. Existing credit rating models rely on financial metrics and deep learning. However, they often overlook insights from non-financial data, such as corporate annual reports. To address this, this paper introduces a corporate credit rating framework that integrates financial data with features extracted from annual reports using FinBERT, aiming to fully leverage the potential value of unstructured text data. In addition, we have developed a large-scale dataset, the Comprehensive Corporate Rating Dataset (CCRD), which combines both traditional financial data and textual data from annual reports. The experimental results show that the proposed method improves the accuracy of the rating predictions by 8-12%, significantly improving the effectiveness and reliability of corporate credit ratings.

Problem

• Integrates financial data and annual report text for credit rating
• Improves rating prediction accuracy by 8-12%
• Creates a large-scale dataset combining financial and textual data

Innovation

• Integrates financial data with FinBERT-extracted annual report features
• Develops Comprehensive Corporate Rating Dataset (CCRD)
• Improves rating prediction accuracy by 8-12%