From Accuracy to Auditability: A Survey of Determinism in Financial AI Systems

• URL: http://arxiv.org/abs/2605.23955v2
• Date: Wed, 27 May 2026 17:26:09 GMT
• Title: From Accuracy to Auditability: A Survey of Determinism in Financial AI Systems
• Abstract summary: Machine learning in regulated financial environments exposes critical vulnerabilities in algorithmic vulnerabilities.<n>Deep neural networks and Generative AI have introduced mechanical nondeterminism rooted in hardware and architecture.<n>This survey provides a systems perspective on failures across three modalities now dominant in financial AI.
• Score: 16.784964226963744
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Deploying machine learning in regulated financial environments -- credit risk, fraud detection, and anti-money laundering -- exposes critical vulnerabilities in algorithmic reproducibility. While early financial ML addressed statistical challenges such as backtest overfitting, deep neural networks and Generative AI have introduced mechanical nondeterminism rooted in hardware and architecture. This survey provides a systems perspective on reproducibility failures across three modalities now dominant in financial AI: tabular models (post-hoc explanation variance), graph networks (stochastic sampling and temporal asynchrony), and LLM-based agentic workflows (batch-dependent divergence and trajectory drift). We supplement the literature analysis with first-party experiments on public financial datasets -- quantifying explanation rank instability in credit scoring, prediction flip rates in GNN-based fraud detection, and tensor-parallel-induced output divergence in LLM entity extraction. We propose a layered evaluation framework linking modality-specific metrics (RBO, D_cos, TDI, PSD) to audit readiness, and empirically validate the complementarity of logit-level and semantic-level determinism measures.

Related papers

• AI Slop and Hallucinations in Vulnerability Assessment: A Survey on Reasoning Failures and Trustworthy Mitigation [48.55515767840701]
  "AI slop" is artifacts, hallucinated vulnerabilities, plausible but incorrect patches, and semantically repackaged bug reports.<n>This paper surveys the empirical evidence, identifies a unifying mechanism, and traces a path toward trustworthy triage.
  arXiv  Detail & Related papers  (2026-08-26T11:48:38Z)
• When AI Meets Wall Street: A Survey on Trustworthy AI in Fintech [32.022055126183865]
  We propose a unified, lifecycle-centric and mechanism-driven framework for financial AI.<n>We partition financial AI into three lifecycle stages: training and updating, deployment and inference, and operation, monitoring, and feedback.<n>For each subtype, we analyse algorithmic strategy, feasibility constraints, stealth and persistence, and downstream financial consequences.
  arXiv  Detail & Related papers  (2026-05-28T23:10:04Z)
• VerifyMAS: Hypothesis Verification for Failure Attribution in LLM Multi-Agent Systems [79.51005192758262]
  Large language model-driven multi-agent systems excel at complex tasks.<n>However, unreliable agents remain a key bottleneck to system-level reliability.<n>We propose a hypothesis verification framework for agent failure attribution.
  arXiv  Detail & Related papers  (2026-05-17T14:09:35Z)
• Hybrid LSTM-Graph Neural Framework for Robust Financial Fraud Detection and Adversarial Resilience [0.0]
  This paper proposes FraudShield AI, a hybrid framework that integrates Long Short-Term Memory (LSTM) networks with hand-crafted Graph Topological Features.<n>A Focal Loss objective is used to address class imbalance, and a dynamic thresholding mechanism is introduced to improve resilience against low-value smurfing attacks.
  arXiv  Detail & Related papers  (2026-05-02T02:10:20Z)
• Fraud Detection System for Banking Transactions [0.0]
  This research introduces a machine learning-based fraud detection framework utilizing the PaySim synthetic financial transaction dataset.<n>The proposed framework provides a robust and scalable solution to enhance fraud prevention capabilities in FinTech transaction systems.
  arXiv  Detail & Related papers  (2026-04-09T08:17:27Z)
• CAMD: Coverage-Aware Multimodal Decoding for Efficient Reasoning of Multimodal Large Language Models [48.33850939857255]
  We develop a theoretical framework that links sampling coverage, instance difficulty, and residual risk.<n>Our analysis reveals that multimodal reasoning exhibits a heavy-tailed difficulty distribution.<n>We propose Coverage-Aware Multimodal Decoding (CAMD), an adaptive inference mechanism that dynamically allocates according to estimated uncertainty.
  arXiv  Detail & Related papers  (2026-03-16T02:31:03Z)
• ProbeLLM: Automating Principled Diagnosis of LLM Failures [89.44131968886184]
  We propose ProbeLLM, a benchmark-agnostic automated probing framework that elevates weakness discovery from individual failures to structured failure modes.<n>By restricting probing to verifiable test cases and leveraging tool-augmented generation and verification, ProbeLLM grounds failure discovery in reliable evidence.
  arXiv  Detail & Related papers  (2026-02-13T14:33:13Z)
• STAR : Bridging Statistical and Agentic Reasoning for Large Model Performance Prediction [78.0692157478247]
  We propose STAR, a framework that bridges data-driven STatistical expectations with knowledge-driven Agentic Reasoning.<n>We show that STAR consistently outperforms all baselines on both score-based and rank-based metrics.
  arXiv  Detail & Related papers  (2026-02-12T16:30:07Z)
• ProteuS: A Generative Approach for Simulating Concept Drift in Financial Markets [44.76567557906836]
  A fundamental problem in developing and validating adaptive algorithms is the lack of a ground truth in real-world financial data.<n>This paper introduces a novel framework, named ProteuS, for generating semi-synthetic financial time series with pre-defined structural breaks.<n>An analysis of the generated data confirms the complexity of the task, revealing significant overlap between the different market states.
  arXiv  Detail & Related papers  (2025-08-30T21:01:47Z)
• Matrix Factorization for Inferring Associations and Missing Links [5.700773330654261]
  Missing link prediction identifies unseen but potentially existing connections in a network.<n>In proliferation detection, this supports efforts to identify and characterize attempts by state and non-state actors to acquire nuclear weapons.<n>We introduce novel weighted (WNMFk), Boolean (BNMFk), and Recommender (RNMFk) matrix factorization methods, along with ensemble variants incorporating logistic factorization, for link prediction.
  arXiv  Detail & Related papers  (2025-03-06T18:22:46Z)
• The Misclassification Likelihood Matrix: Some Classes Are More Likely To Be Misclassified Than Others [1.654278807602897]
  This study introduces Misclassification Likelihood Matrix (MLM) as a novel tool for quantifying the reliability of neural network predictions under distribution shifts. The implications of this work extend beyond image classification, with ongoing applications in autonomous systems, such as self-driving cars.
  arXiv  Detail & Related papers  (2024-07-10T16:43:14Z)
• Advancing Anomaly Detection: Non-Semantic Financial Data Encoding with LLMs [49.57641083688934]
  We introduce a novel approach to anomaly detection in financial data using Large Language Models (LLMs) embeddings. Our experiments demonstrate that LLMs contribute valuable information to anomaly detection as our models outperform the baselines.
  arXiv  Detail & Related papers  (2024-06-05T20:19:09Z)
• Bayesian Bilinear Neural Network for Predicting the Mid-price Dynamics in Limit-Order Book Markets [84.90242084523565]
  Traditional time-series econometric methods often appear incapable of capturing the true complexity of the multi-level interactions driving the price dynamics. By adopting a state-of-the-art second-order optimization algorithm, we train a Bayesian bilinear neural network with temporal attention. By addressing the use of predictive distributions to analyze errors and uncertainties associated with the estimated parameters and model forecasts, we thoroughly compare our Bayesian model with traditional ML alternatives.
  arXiv  Detail & Related papers  (2022-03-07T18:59:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.