Research

Paper

AI LLM March 12, 2026

Modal Logical Neural Networks for Financial AI

Authors

Antonin Sulc

Abstract

The financial industry faces a critical dichotomy in AI adoption: deep learning often delivers strong empirical performance, while symbolic logic offers interpretability and rule adherence expected in regulated settings. We use Modal Logical Neural Networks (MLNNs) as a bridge between these worlds, integrating Kripke semantics into neural architectures to enable differentiable reasoning about necessity, possibility, time, and knowledge. We illustrate MLNNs as a differentiable ``Logic Layer'' for finance by mapping core components, Necessity Neurons ($\Box$) and Learnable Accessibility ($A_θ$), to regulatory guardrails, market stress testing, and collusion detection. Four case studies show how MLNN-style constraints can promote compliance in trading agents, help recover latent trust networks for market surveillance, encourage robustness under stress scenarios, and distinguish statistical belief from verified knowledge to help mitigate robo-advisory hallucinations.

Metadata

arXiv ID: 2603.12487
Provider: ARXIV
Primary Category: cs.LG
Published: 2026-03-12
Fetched: 2026-03-16 06:01

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.12487v1</id>\n    <title>Modal Logical Neural Networks for Financial AI</title>\n    <updated>2026-03-12T22:12:18Z</updated>\n    <link href='https://arxiv.org/abs/2603.12487v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.12487v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>The financial industry faces a critical dichotomy in AI adoption: deep learning often delivers strong empirical performance, while symbolic logic offers interpretability and rule adherence expected in regulated settings. We use Modal Logical Neural Networks (MLNNs) as a bridge between these worlds, integrating Kripke semantics into neural architectures to enable differentiable reasoning about necessity, possibility, time, and knowledge. We illustrate MLNNs as a differentiable ``Logic Layer'' for finance by mapping core components, Necessity Neurons ($\\Box$) and Learnable Accessibility ($A_θ$), to regulatory guardrails, market stress testing, and collusion detection. Four case studies show how MLNN-style constraints can promote compliance in trading agents, help recover latent trust networks for market surveillance, encourage robustness under stress scenarios, and distinguish statistical belief from verified knowledge to help mitigate robo-advisory hallucinations.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <published>2026-03-12T22:12:18Z</published>\n    <arxiv:comment>4 pages, 1 figure, Accepted at ICLR 2026 FinAI</arxiv:comment>\n    <arxiv:primary_category term='cs.LG'/>\n    <author>\n      <name>Antonin Sulc</name>\n    </author>\n  </entry>"
}