Explainability

2 items

DOCDEV.to AI·13d ago

Building AI Agents for Compliance Monitoring in Finance: Architecture That Passes Auditors

This content discusses building AI agents for compliance monitoring in finance, focusing on architectures that provide explainable decisions to satisfy auditors and regulators. It highlights that explainability is a regulatory requirement, not a nice-to-have, and explains how to design systems that document reasoning for automated compliance decisions.

Finance Explainability learning compliance

RESEARCHarXiv CS.AI·4/21/2026

The Query Channel: Information-Theoretic Limits of Masking-Based Explanations

This paper formulates masking-based AI explanation methods as communication over a query channel, where explanations act as messages. It derives information-theoretic limits for the recovery of exact explanations, showing that reliable recovery is achievable below a certain capacity.

information theory AI models Explainability feature importance