From Accuracy to Auditability: A Survey of Determinism in Financial AI Systems
The paper titled 'From Accuracy to Auditability: A Survey of Determinism in Financial AI Systems' explores the challenges of deploying machine learning in regulated financial environments. It highlights issues of algorithmic reproducibility and the impact of deep learning technologies on determinism. The authors propose a framework for evaluating audit readiness in financial AI systems based on various metrics.
- ▪The survey addresses vulnerabilities in algorithmic reproducibility in financial machine learning applications.
- ▪It examines three dominant modalities in financial AI: tabular models, graph networks, and LLM-based workflows.
- ▪The authors conducted experiments on public financial datasets to quantify explanation instability and prediction divergence.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2605.23955 (cs) [Submitted on 11 May 2026] Title:From Accuracy to Auditability: A Survey of Determinism in Financial AI Systems Authors:Ruizhe Zhou, Xiaoyang Liu, Gaoyuan Du, Yi Zheng, Shouxi Ren, Deepayan Chakrabarti, Dengdu Jiang View a PDF of the paper titled From Accuracy to Auditability: A Survey of Determinism in Financial AI Systems, by Ruizhe Zhou and 6 other authors View PDF HTML (experimental) Abstract:Deploying machine learning in regulated financial environments -- credit risk, fraud detection, and anti-money laundering -- exposes critical vulnerabilities in algorithmic reproducibility.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
