---
title: Rolling Forward: Enhancing LightGCN with Causal Graph Convolution for Credit Bond Recommendation
url: https://www.ml-quant.com/papers/doi/10-1145-3677052-3698683/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: doi:10.1145/3677052.3698683
source_url: http://dx.doi.org/10.1145/3677052.3698683
featured: 2025-03-20
citations: 2
topic: Risk, Credit & Banking
---


# Rolling Forward: Enhancing LightGCN with Causal Graph Convolution for Credit Bond Recommendation

The study presents an improved version of the Light Graph Convolutional Network that learns over time, enhancing its performance in time-sensitive applications, especially in recommending financial products.

- Source: http://dx.doi.org/10.1145/3677052.3698683
- Identifier: doi:10.1145/3677052.3698683
- Released: 2025-03-18
- First featured: Quant Letter No. 89 (2025-03-20): https://www.ml-quant.com/issues/2025-03-20/
- Citations (Semantic Scholar): 2
- Published in: Proceedings of the 5th ACM International Conference on AI in Finance
- Topic: Risk, Credit & Banking

## Related

- [Unleashing the power of text for credit default prediction: Comparing human-written and generative AI-refined texts](https://www.ml-quant.com/papers/arxiv/2503.18029/): The study shows that using AI language model, ChatGPT, in lending decisions can improve credit default predictions and increase profitability in finance.
- [XVA modelling: validation, performance and model risk management](https://www.ml-quant.com/papers/doi/10-1007-s10479-023-05323-4/): The article explores the impact of XVA on derivatives pricing, emphasizing its model risk and computational effort, and offers a guide for creating a strong model for collateralized exposure and XVA.
- [A cost of capital approach to determining the LGD discount rate](https://www.ml-quant.com/papers/arxiv/2503.23992/): The paper introduces a new method for estimating banking losses, using a market-consistent price for defaulted loans.
- [DEEP LEARNING FOR FINANCIAL STRESS TESTING: A DATA-DRIVEN APPROACH TO RISK MANAGEMENT](https://www.ml-quant.com/papers/ssrn/5146509/): A new deep learning-based framework for financial stress testing is introduced, combining financial indicators to improve risk prediction accuracy and reduce financial risks.
- [The Effect of Financial Risks on the Financial Performance of Banks listed on Bahrain Bourse: An Empirical Stud](https://www.ml-quant.com/papers/ssrn/5147102/): Capital risks significantly impact the financial performance of banks listed on the Bahrain Bourse.
- [Multi-Layer Deep xVA: Structural Credit Models, Measure Changes and Convergence Analysis](https://www.ml-quant.com/papers/arxiv/2502.14766/): The article suggests a new model for portfolio valuation adjustments that uses a deep BSDE approach to handle each layer separately, making it more computationally efficient and adaptable to complex portfolios.