arXiv:2607.06407v1 Announce Type: new Abstract: The XAI community has studied a wide range of queries and scores for explaining predictions of ML models. From a data management perspective, this proliferation of explanation notions calls for declarative query languages in which such notions can be specified, combined, and analyzed uniformly. In this paper, we develop such a framework for Boolean models. We first revisit FOIL, an interpretability query language for black-box models, and show that it has two fundamental limitations: it cannot express central optimality-based explanation queries,
The proliferation of various explanation methods in XAI necessitates a more unified and declarative approach to manage their complexity, making this research timely.
A declarative query language for AI explanations enhances the transparency and reliability of classification models, crucial for trust and widespread adoption in critical applications.
This research introduces a novel framework that allows for more systematic and scalable generation and analysis of explanations for AI models, moving beyond ad-hoc methods.
- · AI developers
- · Regulatory bodies
- · Industries relying on AI (e.g., finance, healthcare)
- · Proprietary, opaque AI explanation tools
- · Developers resistant to transparent AI
Improved debugging and auditing capabilities for complex AI systems.
Increased adoption of AI in highly regulated sectors due to enhanced interpretability.
Potential for new standards and regulations around AI explainability driven by common declarative frameworks.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI