Paper Overview
- Research area: Machine Learning
- Authors: Marcin Kostrzewa, Sebastian Tomczak, Roman Furman
- Published: 2025-05-09
- arXiv: 2505.07227
- Introduces a large-scale public benchmark for corporate bankruptcy prediction covering 2006–2021 in the V4 economies.
- Provides over 1,000,000 company-year records, substantially larger than existing free benchmarks.
- Targets the twin challenges of severe class imbalance and multi-horizon forecasting.
- Enables systematic benchmarking of tabular foundation models, LLMs, and conventional ML approaches.
- Paper: arXiv:2505.07227
Summary
Corporate bankruptcy prediction is a high-stakes financial task characterized by severe class imbalance and multi-horizon forecasting demands. Public datasets supporting it remain scarce and small: widely used free benchmarks contain between 6,000 and 80,000 company-year observations, while larger resources are behind subscription paywalls.
To address this gap, the authors introduce V4FinBench, a benchmark of over one million company-year records from the Visegrád Group (V4) economies (2006–2021). The benchmark is designed to compare tabular foundation models, large language models (LLMs), and standard machine learning methods on bankruptcy prediction under realistic conditions of extreme class imbalance and multi-horizon forecasting requirements.